Valuation is a central component of many high-stakes commercial disputes, from shareholder appraisal actions and bankruptcy proceedings to tax controversies and breach-of-contract claims. In these contexts, the determination of a company’s fair market value often hinges on the present value of projected cash flows, making the discounting exercise a critical step in the valuation process.

In financial modeling, particularly when applying discounted cash flow (“DCF”) methods to estimate the value of a business or an asset, the discount rate plays a pivotal role in translating future cash flows into a present value.^[1] Traditionally, practitioners apply a single discount rate to all future cash flows, reflecting the overall risk profile and the time value of money for the investment. In certain cases, however, the use of multiple discount rates may be considered more appropriate, particularly when future cash flows differ meaningfully in terms of risk, timing, or underlying economic conditions.

For instance, when a company or asset’s risk profile is expected to remain relatively stable over time, applying a single discount rate may provide a reasonable approximation of value. In contrast, when risks evolve materially across different phases of a business, such as in industries with distinct development and commercialization stages, a framework of multiple discount rates may better capture these changing risk characteristics.^[2]

Because discounting compresses a range of future expectations into a single present value figure, even modest variations in the choice of methodology can lead to significant differences in the resulting valuation. The effect of discount rate selection on valuation can be particularly pronounced in litigation involving large, complex businesses, where valuations may reach into the billions of dollars.

Consequently, the choice between using a single discount rate or multiple rates is not merely a technical modeling decision; it can be a focal point of dispute, with significant legal and financial implications. Understanding the considerations that motivate each approach is therefore critical for litigators, experts, and courts as they assess the credibility and robustness of competing valuation analyses.

Discounted Cash Flow in Valuation Litigation

The DCF method is a commonly used tool in financial valuation and is frequently applied in both corporate finance and investment analysis. DCF estimates the present value of a business or asset by projecting its expected future cash flows and discounting those cash flows back to a selected point in time using an appropriate discount rate. The discount rate reflects the opportunity cost of capital (i.e., what investors could earn in an alternative investment of comparable risk) and incorporates both the time value of money and the uncertainty associated with receiving future cash flows.^[3] It can be influenced by factors such as macroeconomic conditions, capital structure, and exposure to market risk.

The DCF model is widely accepted in enterprise valuation practice. It focuses on a business’s expected future performance, which is particularly important for companies with unique capital structures, limited public comparables, or evolving business models to consider. These company characteristics make DCF especially relevant in valuation-related litigation, where generating a case-specific, independent estimate of value is often paramount. For example, courts have relied on DCF in shareholder appraisal actions,^[4] breach of fiduciary duty claims,^[5] lost profits assessments,^[6] and bankruptcy proceedings.^[7]

DCF is among the valuation methods that have been considered by courts, including the Delaware Court of Chancery. For example, in In re Appraisal of Dell Inc., the Court emphasized that “the DCF . . . methodology has featured prominently in this Court because it is the approach that merits the greatest confidence within the financial community.”^[8] Similarly, in Andaloro v. PFPC Worldwide, the Court stated that DCF “is frequently used in the Delaware Chancery Court and many prefer to give it great, and sometimes even exclusive, weight when it may be used responsibly.”^[9]

A typical DCF analysis in enterprise valuation involves three key components: (1) projecting cash flows over a discrete forecast period, (2) estimating a terminal value to capture the value of cash flows beyond that horizon, and (3) applying a discount rate to convert future cash flows into present value. Each of these components can significantly affect the outcome and is often subject to dispute in litigation. Differences in projections, assumptions about terminal growth, or the construction of the discount rate can lead to materially different conclusions about value.

DCF valuations are sensitive to the selection of the discount rate and the methodology used to apply it. Because discounting compresses future cash flows into a single present value figure, small changes in the discount rate—or in the way rates are applied across time—can result in large changes in the final valuation. This sensitivity becomes especially consequential in high-stakes litigation, where valuations often involve complex, multi-billion-dollar businesses. The discussion that follows examines the considerations underlying the choice between a single and multiple discount rates approach, as well as the potential valuation consequences in litigation settings.

Single Discount Rate Approach

The single discount rate approach is the most common method in Discounted Cash Flow analysis. Under this approach, all projected future cash flows are discounted back to a particular time using a constant rate that reflects the overall risk profile of the business or asset. This rate captures both the time value of money and the risk associated with receiving those cash flows over time.

In practice, when valuing a firm’s free cash flows, the single discount rate is typically based on the firm’s weighted average cost of capital (“WACC”), which represents the average return required by all capital providers—both debt and equity. The capital asset pricing model (“CAPM”) is commonly used to price the required return to equity. CAPM calculates the cost of equity as the sum of the risk-free rate (the return on a virtually riskless investment, such as a U.S. Treasury bond), the market risk premium (the additional return expected from investing in the overall market above the risk-free rate), and the company’s beta (a measure of how sensitive the company’s equity returns are to movements in the overall market). A beta greater than one indicates higher systematic (undiversifiable) risk than the market, while a beta below one suggests lower market-related risk.

The cost of debt, by contrast, reflects the effective rate the firm pays on its borrowings, adjusted for the tax deductibility of interest. In practice, various approaches are used to estimate it, including using the firm’s current yield on debt or the sum of a risk-free rate and a credit spread that captures the firm’s default risk. The WACC combines these two components—the cost of equity and the after-tax cost of debt—weighted by their respective proportions in the firm’s capital structure.

Once determined, this single discount rate is applied to all projected cash flows, regardless of when they occur. This creates a consistent framework for evaluating a stream of future benefits based on a single measure of opportunity cost and risk exposure.^[10]

To illustrate, consider a simplified model of a business that expects to generate $10 million in free cash flows annually for the next ten years, and assume the firm’s WACC is estimated at 12%.^[11] The present value of this future cash flow stream is calculated as

This means the total present value of the ten-year $10 million annual cash flow stream, when discounted at a constant 12% rate, is approximately $56.5 million.

The single discount rate approach is often used in valuation models due to its conceptual clarity and ease of application, particularly when a firm’s risk profile and capital structure are expected to remain relatively stable over time. In cases where those factors vary meaningfully, however, applying a constant rate across all cash flows may not fully reflect the underlying economic characteristics of the business.

Multiple Discount Rates Approach

Traditional DCF models often apply a single discount rate to all projected cash flows, but a body of academic literature has raised concerns about the limitations of this assumption. Researchers have pointed to various settings where a constant discount rate may not reflect the economic characteristics of the underlying investment.^[12]

First, several studies suggest that the required rate of return used in valuation models can vary over time due to changing economic conditions and the evolving opportunity cost of capital—that is, the return investors forgo by allocating funds to a specific project rather than to alternative investments. In such cases, the assumption of a constant discount rate may lead to valuation results that do not fully capture the time-varying risk profile of the investment.^[13]

Second, some researchers argue that many investment projects exhibit nonconstant risk exposure over their life cycle, with distinct phases that carry different types and magnitudes of risk. For instance, a project may begin with high levels of idiosyncratic or technical uncertainty and later transition into phases that are more exposed to systematic market risks. In such settings, applying a single discount rate across all cash flows could produce misleading valuation outcomes, as the method does not reflect the underlying changes in project risk.^[14]

A related line of literature has focused on component-based risk modeling, where the risk of individual cash flow components is evaluated separately. This approach aims to assess risk by framing it in terms of fair insurance premiums—that is, what an investor would pay to insure against the variability or potential loss of specific cash flows.^[15]

In certain industries, operational phases may be more distinctly associated with different risk profiles. For example, in the oil and gas industry, early-stage activities such as exploration and drilling are often subject to technical, regulatory, or geological uncertainties that may not be closely correlated with broad market fluctuations.^[16] In contrast, later stages of production, processing, and distribution are more directly affected by market-driven factors such as commodity price volatility, global demand, and macroeconomic trends.

In the biologics and pharmaceutical sectors, early stages of drug development, such as preclinical research and clinical trials, may involve substantial project-specific uncertainty.^[17] These risks are often idiosyncratic in nature, stemming from scientific or operational challenges unique to a given project.

From a corporate finance perspective, such risks may be better reflected through adjustments to expected cash flows, for example, by incorporating probability-weighted outcomes, rather than through changes to the discount rate.^[18] As a product advances toward commercialization, its cash flows may become increasingly sensitive to more systemic risks, such as pricing pressure, reimbursement policy, and competitive dynamics, which can, in turn, be captured through the discount rate. In these types of multi-phase businesses, some valuation practitioners have examined whether differentiated discount rates could better reflect the risk characteristics across time.^[19]

A recent decision from the Delaware Court of Chancery illustrates that courts, too, recognize that it can be appropriate to apply different discount rates to cash flows with distinct risk characteristics. In Shareholder Representative Services LLC v. Alexion Pharmaceuticals, Inc.,^[20] the Court applied a lower, debt-based rate for development and regulatory milestones,^[21] and a higher, equity-based rate for a sales-linked milestone exposed to market risk,^[22] recognizing that discount rates should align with the nature of the underlying uncertainty.

In practice, implementing a DCF model with multiple discount rates involves a structured and analytically grounded process. First, the projected cash flows are segmented into distinct components or phases based on their timing, source, or nature, such as R&D expenditures, operational revenues, or milestone payments. For each segment, the associated risk characteristics must be identified, taking into account factors such as macroeconomic conditions, industry-specific volatility, regulatory uncertainty, market exposure, and the degree of diversification.

Second, based on this risk assessment, a distinct discount rate is assigned to each cash flow segment. These rates may reflect varying costs of capital, degrees of market correlation, or differing investor return expectations, depending on the nature of the risk. Finally, once each component has been matched with a corresponding discount rate, the present value of each cash flow is calculated using standard DCF mechanics, and the results are aggregated to arrive at the overall valuation.

The following example illustrates a simplified application of how a multiple discount rates method may be used in practice. Consider a hypothetical firm with the following projected cash flows over a ten-year period:

• Years one to five: Negative $10 (–$10) million per year (e.g., investment in R&D with idiosyncratic, diversifiable risks)
• Years six to ten: +$25 million per year (e.g., commercialization phase with market risks)

Under a multiple discount rates approach, one might apply a lower rate (e.g., the risk-free rate of 4% in this example) to the first phase and a higher rate (e.g., a 12% WACC) to the second phase.^[23] The present value calculation would be:

Consequences of DCF Assumptions in Valuation

The following examples illustrate the practical implications of selecting a discounting methodology in a DCF analysis. The first example demonstrates how the use of a single versus multiple discount rates can result in materially different valuation outcomes when applied to the same set of projected cash flows. The second example shows that it is possible to produce equivalent valuation results using either method; however, achieving such equivalence may require selecting a single discount rate that lacks clear theoretical or empirical support.

Together, these underscore the importance of critically assessing both methodological choices and input assumptions in valuation exercises, particularly in high-stakes litigation contexts where even small modeling decisions may have significant financial implications.

Example 1

Consider the same hypothetical scenario presented in the previous section, with a firm with projected cash flows over a ten-year period, consisting of negative $10 million per year in years one through five, and positive $25 million per year in years six through ten.

As shown in the previous section, the present value of this cash flow stream under a multiple discount rates approach, where the first five years are discounted at a risk-free rate of 4%, and the second five years are discounted at a WACC of 12%, is approximately $6.6 million.

In practical applications, valuation analysts may apply a company’s WACC as a single discount rate across all projected cash flows, implicitly assuming a uniform level of risk throughout the forecast period. In the present example, this translates into applying a 12% discount rate to both the initial five years of negative cash flows and the subsequent five years of positive cash flows. This treatment assumes that all cash flows—regardless of their underlying risk characteristics—should be discounted at the same rate:

The resulting valuation outcome differs substantially from that of the multiple discount rates approach, producing a materially higher present value by a factor of nearly 2.3,^[24] despite identical cash flow projections.

Figure 1 below shows the year-by-year comparison of the two methods. The main difference between them is that the multiple discount rates method applies a smaller discount to early losses, whereas the single discount rate applies a much higher rate of 12% to those same negative cash flows—resulting in greater net negative cash flows (in absolute terms) for the former during the early years.

Figure 1: DCF with Single and Multiple Discount Rates Resulting in Different Valuation Outcomes

This highlights the extent to which a methodological choice can influence valuation results. In litigation, such as matters involving damages, shareholder disputes, or bankruptcy proceedings, the choice of discount rates may materially affect outcomes related to liability, compensation, or settlement terms. The example illustrates that even when cash flow projections are held constant, the decision to apply a single or multiple discount rates framework may have outsized effects on the final valuation.

Example 2

In this second scenario, we revisit the same projected cash flows but ask a different question: Is there a single discount rate that produces the same present value as the multiple discount rates model? The answer is yes. By adjusting the discount rate applied uniformly across all ten years, one can identify a rate—approximately 15.8%—that equates the present value of the cash flows to the $6.6 million obtained under the multiple rate approach.

Figure 2 below shows a comparison of the two methods. Although discounting with a single rate of 15.8% results in smaller net negative cash flows in absolute terms during the early loss-making years, the multiple discount rates approach catches up in the later period of positive earnings, as it applies a lower rate of 12% to those cash flows—ultimately resulting in equivalent net cash flows.

Figure 2: DCF with Single and Multiple Discount Rates Resulting in Identical Valuation Outcomes

While this may appear to reconcile the two methodologies in terms of numerical outcome, it raises a conceptual challenge: In this example, the 15.8% rate is neither the firm’s hypothetical cost of capital of 12% nor the risk-free rate of 4%. In fact, it lies outside the range defined by those two benchmarks. It may not reflect any identifiable investment alternative, capital market benchmark, or macroeconomic condition. In other words, while mathematically convenient, the adjusted discount rate here lacks a clear theoretical or empirical foundation, which may weaken its credibility in expert analysis or legal proceedings.

This example reinforces that numerical equivalence does not necessarily guarantee conceptual soundness. In practice, the selection of discount rates, whether single or multiple, should be supported by economic rationale, consistency with market expectations, and alignment with the risks underlying the projected cash flows.

Conclusion

Valuation plays a critical role in many litigation settings, where financial conclusions can influence outcomes related to liability, damages, and settlement terms. DCF is one of the more frequently applied methodologies in these contexts of enterprise valuation, due to its flexibility and grounding in forward-looking projections. Within this framework, the decision to apply a single or multiple discount rates can have a meaningful impact on the resulting valuation.

As demonstrated in the examples, both approaches can produce significantly different outcomes, even when based on the same projected cash flows. While it is sometimes possible to align their numerical results, doing so may require selecting a discount rate that lacks a clear theoretical or empirical reference point. This underscores the importance of articulating the rationale behind discount rate assumptions and ensuring consistency with the underlying risk characteristics of the cash flows.

Ultimately, the appropriate methodology will depend on the specific facts of the case, the nature of the business or asset being valued, and the analytical goals of the valuation exercise. Maintaining transparency in modeling decisions and grounding assumptions in sound economic reasoning are essential across all approaches.

1. While the discount rate is a critical input, and this article specifically focuses on assumptions related to discount rate selection, other factors also significantly influence the DCF outcome. These include, for example, the length of the projected cash flow period, the size and timing of expected cash flows, the timing of discounting, the assumptions used for terminal value, the growth rates applied during the forecast period (which reflect expectations about how the business or asset will grow over time), and the overall risk profile of the asset or business being evaluated. ↑

2. U.S. courts have recognized this distinction and adopted multiple-rate approaches where appropriate to reflect phase-specific risk dynamics. See, e.g., S’holder Representative Servs. LLC v. Alexion Pharms., Inc., 341 A.3d 513 (Del. Ch. 2025). ↑

3. See Richard A. Brealey, Stewart C. Myers & Franklin Allen, Principles of Corporate Finance, 20–45 (Ch. 2: How to Calculate Present Values) (13th ed., McGraw-Hill Education, 2020); Jonathan Berk & Peter DeMarzo, Corporate Finance 246–272 (Ch. 7: Investment Decision Rules) (5th ed., Pearson 2020). ↑

4. See, e.g., Northwest Inv. Corp. v. Wallace, 741 N.W.2d 782 (Iowa 2007). ↑

5. See, e.g., Su v. Bensen, No. 19-3178, 2024 U.S. Dist. LEXIS 145404 (D. Ariz. Aug. 15, 2024). ↑

6. See, e.g., BP Amoco v. Flint Hills Res., No. 05 C 5661, 2009 U.S. Dist. LEXIS 131272 (N.D. Ill. June 4, 2009). ↑

7. See, e.g., Dietz v. Jacobs, No. 12-1628, 2014 U.S. Dist. LEXIS 37144 (D. Minn. Mar. 21, 2014). ↑

8. In re Appraisal of Dell Inc. No. 9322-VCL, 2016 Del. Ch. LEXIS 81 (Del. Ch. May 31, 2016). ↑

9. Andaloro v. PFPC Worldwide, Nos. 20336, 20289, 2005 Del. Ch. LEXIS 125 (Del Ch. Aug. 19, 2005). ↑

10. This approach is generally more appropriate when a firm undertakes a project consistent with its core operations. For projects that fall outside the firm’s typical business (for example, a utility company investing in a tech startup), the risk profile may differ significantly. In such cases, one may prefer to use a discount rate different from the firm’s overall WACC. ↑

11. All discounting examples presented in this article assume that cash flows are realized at the end of each year (e.g., t = 1 for year 1, t = 2 for year 2, and so on) and are discounted back to the beginning of the hypothetical study period (i.e., t = 0). ↑

12. Note that these studies highlight ongoing discussion around the limitations of single-rate DCF models and the potential for alternative methods, such as multi-rate discounting, to address evolving or segmented risk structures within an investment or project. Whether and how these ideas are implemented in practice varies, and their relevance depends on the specific assumptions and characteristics of the valuation at hand. ↑

13. Geltner and Mei (1995) and Tiwari (1994) show that shifts in interest rates, inflation, and market risk premiums can cause required returns to fluctuate, making the assumption of a constant discount rate unrealistic. Investors, therefore, form expectations at t = 0 not only about future cash flows but also about how discount rates may change as market conditions evolve. See Kashi Nath Tiwari, Single Versus Multiple Discount Rates in Investment Theory, 7 J. Financial & Strategic Decisions 19–42 (1994); David Geltner & Jianping Mei, The Present Value Model with Time-Varying Discount Rates: Implications for Commercial Property Valuation and Investment Decisions, 11 J. Real Estate Fin. & Econ. 119–135 (1995). Empirical evidence further indicates that valuation models incorporating time-varying discount rates better explain observed market values than those assuming a single, constant rate. See Sujata Behera, Does the EVA Valuation Model Explain the Market Value of Equity Better Under Changing Required Return Than Constant Required Return?, 6 Fin. Innovation 9 (2020). ↑

14. See Vicente Alcaraz, Should Practitioners (Continue to) Use a Single Discount Rate in Large-Scale Project Valuation?, 20 J. Structured Fin. 93 (2014); Ken Fuller et al., The Problem of Discount Rate in Infrastructure Projects: Exploring the Idea of Financial Twins (Feb. 6, 2023) (unpublished manuscript) (SSRN); Babak Jafarizadeh & Reidar B. Bratvold, Project Economics in the Big-Bets Industry: The Integrated Valuation in Practice, 197 J. Petroleum Sci. & Eng’g 108095 (2021). ↑

15. See Brealey, Myers & Allen, Principles of Corporate Finance 590–613 (Ch. 22: Real Options); Avinash K. Dixit & Robert S. Pindyck, Investment Under Uncertainty (1st ed. Princeton University Press 1994); David Espinoza et al., DNPV: A Valuation Methodology for Infrastructure and Capital Investments Consistent with Prospect Theory, 38 Constr. Mgmt. & Econ. 259–274 (2020). ↑

16. See, e.g., Brealey, Myers & Allen, Principles of Corporate Finance 228–256 (Ch. 9: Risk and the Cost of Capital), which lists the risk of a dry hole in oil exploration as an example of “bad outcome” that appears to reflect diversifiable risks that would not affect the expected rate of return demanded by investors. ↑

17. See Brealey, Myers & Allen, Principles of Corporate Finance Ch. 9, which lists the risk of unacceptable side effects of a new drug as an example of “bad outcome” that appears to reflect diversifiable risks that would not affect the expected rate of return demanded by investors. ↑

18. See Brealey, Myers & Allen, Principles of Corporate Finance 260 (“Diversifiable risks do not increase the cost of capital. They do affect expected cash flows, however.”); Berk & DeMarzo, Corporate Finance 421 (“Firms sometimes try to adjust for this risk by assigning a higher cost of capital to new projects. Such adjustments are generally incorrect, as this execution risk is typically firm-specific risk, which is diversifiable. . . . Of course, this does not mean that we should ignore execution risk. We should capture this risk in the expected cash flows generated by the project.”). ↑

19. These examples are not intended to imply that any particular methodology is preferred or should be used in any specific industry. Rather, they illustrate scenarios in which varying risk exposures over time have led researchers and practitioners to explore valuation methods that account for such differences. ↑

20. S’holder Representative Servs. LLC v. Alexion Pharms., Inc., 341 A.3d 513 (Del. Ch. 2025). ↑

21. See S’holder Representative Servs. LLC v. Alexion Pharms., Inc., 341 A.3d at 547 (“But the risks underlying the metrics in Milestones 2 through 7 are ‘diversifiable.’ A diversifiable risk is one ‘that is peculiar to an individual company.’ Such risks can be ‘diversified away’ through a broad investment portfolio ‘due to the law of large numbers.’ . . . For Milestones 2 through 7, the discount rate is the risk-free rate plus a credit risk premium.” (citations omitted)). ↑

22. See S’holder Representative Servs. LLC v. Alexion Pharms., Inc., 341 A.3d at 548 (“Milestone 8 is different because the risk associated with its underlying metric is ‘nondiversifiable.’ As the name suggests, nondiversifiable (or ‘systemic’) risk, ‘cannot be fully removed through diversification’ because it is ‘correlated with the market.” . . . Milestone 8’s net sales metric carries nondiversifiable risk. So the milestone’s discount rate must therefore include a risk premium.” (citations omitted)). ↑

23. The purpose of this example is to demonstrate how the application of multiple discount rates may affect valuation outcomes under certain assumptions. It does not imply that this approach is preferable, superior, or recommended in any given case. The appropriateness of any methodology depends on factors such as the specific context, assumptions, and purpose of the valuation. ↑

24. $15.1M / $6.6M ≈ 2.3. ↑

Imagine: A compliance officer at a broker-dealer is asked to draft a new written supervisory procedure on reviewing securities transactions of associated persons. The compliance officer asks his personal version of ChatGPT to draft him an appropriate policy. The compliance officer then has a number of interactions with ChatGPT, and it produces a final written supervisory procedure, which is adopted by the firm. Unfortunately, the written supervisory procedure lacked some key elements, and the Financial Industry Regulatory Authority (“FINRA”) asks who wrote the policy. The compliance officer admits that they asked ChatGPT to draft the procedure but failed to retain the prompts sent to ChatGPT or responses received from ChatGPT. FINRA enforcement alleges that the firm failed to keep records and failed to supervise the compliance officer.

Although this is a hypothetical example, it is likely to occur regularly. The promise of generative artificial intelligence is certainly seductive, but there are important issues for broker-dealers to consider. Should the broker-dealer have a policy on employee AI usage and if so, are the prompts and responses records that are required to be retained under Rule 17a-4 of the Exchange Act?

History of the SEC’s Recordkeeping Rule

In order to implement Section 17(a)(1) of the Exchange Act, in 1935 the Securities and Exchange Commission adopted Rules 17a-3 (records to be made) and 17a-4 (records to be preserved). Rule 17a-4(b)(4) still requires a broker-dealer to preserve “[o]riginals of all communications received and copies of all communications sent (and any approvals thereof) by the member, broker or dealer (including inter-office memoranda and communications) relating to its business as such . . . .”^[1]

Though the SEC has modernized Rules 17a-3 and 17a-4 several times over the intervening ninety-one years, it has yet to cleanly advise as to the meaning of “communications” and “business as such.” As a result, compliance officers and lawyers will need to make informed decisions about whether generative AI prompts and responses are communications and whether they relate to the broker-dealer’s business as such.

How Generative AI Work Appears

Most generative AI operates using statistical probability to predict the most likely sequence of tokens (i.e., pieces of data like letters or words) when responding to prompts. If you enter “duck, duck . . .” generative AI will likely respond with the word “goose.” If you were to then prompt “Why goose?” the generative AI would likely respond with something like, “Because goose frequently appears after duck, duck” or explain the game duck, duck, goose. On review, it appears that the person and the AI system are “talking”; some data scientists even refer to the exchange as a “conversation.” This interaction allows the system to refine its outputs based on preceding context, mimicking the flow of natural dialogue between two people in real time. In short, generative AI looks and feels like talking with a person. That look and feel are at the heart of the issue of whether prompts and responses are communications.

Arguments That AI Prompts and Responses Are Not Required Records

There are several good arguments that certain AI prompts and responses should not be treated as 17a-4 records.

• Generative AI Conversations Are Not Communication. In common discourse, communication involves the transmission of information from one person to another person. After all, you don’t talk to a computer; you issue commands. In many respects, prompting is similar to entering a search query into a research database, instructing a spreadsheet to perform a calculation, or directing a word processing program to generate a document, none of which would generally be deemed communications for recordkeeping purposes.
• Many Prompts and Responses Are Not About “Business as Such.” A prompt and response about how to draft written supervisory procedures may not relate to a broker-dealer’s business as such. While they relate to the broker-dealer’s compliance policies, it could be argued that they do not relate to its actual business as such. The phrase “as such” clearly limits the application of Rule 17a-4 in a way that excludes some communications that relate to the broker-dealer’s business tangentially. Here, the AI prompt and response relate to compliance, which could be argued not to be about the trading of securities (i.e., the business as such).^[2]
• Rule 17a-4’s “Inter-Office Memoranda” Parenthetical Confirms the Rule Only Applies to Select One-Sided Communications. If the SEC intended to capture all one-sided communications, it arguably would have said so. Instead, the SEC explicitly referenced one-sided communications in an exclusive way by providing that communications to be saved are “(including inter-office memoranda and communications)” in Rule 17a-4. Because the SEC did not use the phrase “among other things,” inter-office memoranda and communications could be the only one-sided communications required to be retained.

Arguments That AI Prompts and Responses Are Required Records

Conversely, there are several good arguments as to why AI prompts and responses should be treated as 17a-4 records.

• Prompts and Responses Are Indirect Communications with the AI Company’s Employees. Prompts and AI responses could be argued to constitute communication because they are transmitted to and processed by infrastructure controlled by humans. Prompts are sent to corporate entities, which have employees, and responses are received from those same entities and their employees. When a broker-dealer employee prompts a public AI system, that input is not merely an internal mental draft but rather a transmission to an external service provider whose personnel design, maintain, and monitor such inputs, and in some cases review or use them for system improvement or safety.^[3] Moreover, a court recently determined that generative AI usage can invalidate the attorney-client privilege, which would only be possible if there were human viewers.^[4]
• AI Output Has Been Deemed Communication in Another Context. Regulators have already treated correspondence between an AI chatbot and humans as “communications.” FINRA’s Advertising Regulation FAQs make clear that “Firms are responsible for their communications, regardless of whether they are generated by a human or AI technology. Accordingly, firms must ensure that AI-generated communications they distribute or make available comply with applicable federal securities laws and regulations and FINRA rules.”^[5] The FAQs further reflect that communications made available through online or interactive platforms are considered communications with the public if they are accessible to investors, and firms must retain records of such communications in accordance with applicable recordkeeping obligations.^[6] This framework supports the conclusion that prompts submitted to, and outputs generated by, AI or chatbot systems fall within existing communications and recordkeeping requirements without regard for whether there were two humans directly involved in the communication.
• Prompts and Responses Look Like Communication. Prompts and AI-generated responses could be treated as required records because, functionally, they fit squarely within the regulatory concept of a “communication.” The SEC has long emphasized that communications subject to recordkeeping obligations under Rules 17a-3 and 17a-4 include any written or electronic messages relating to a firm’s business, regardless of format or technology.^[7] Furthermore, due to the natural language search and response patterns as well as the ability to interact and train your AI agents, sessions with generative AI are analogous to communication.
• Retention Is Consistent with the Purpose of Rule 17a-3 and 17a-4. The classification of AI prompts and responses as required records hinges on their role as primary source material for a firm’s supervisory framework. Would retaining these provide evidence that could be useful in an enforcement action or examination? Here, the conversations function as the modern equivalent of compliance work papers or internal memoranda. The digital exchanges possess significant probative value as an audit trail revealing the firm’s intent, the depth of its due diligence, and whether the resulting policies were “reasonably designed” to ensure regulatory adherence. Failing to retain these records would undermine the purpose of the recordkeeping requirements, as staff could effectively hide the decision-making process and any potential shortcuts or misunderstandings from regulatory oversight during enforcement action.

The SEC’s Past and Current Positions

Past SEC Chair Gary Gensler brought a series of high-profile enforcement actions against broker-dealers for allowing their staff to use “off-channel” communications methods that created records but did not retain them and for allowing written communications that could not be supervised.^[8] More recently, Commissioner Hester M. Peirce criticized the prior enforcement program as overreaching and as an example of regulation through enforcement.^[9] As a result, under the current administration, it is unlikely that the SEC would bring high-dollar enforcement actions against firms that do not treat their use of generative AI as creating records that must be preserved. However, broker-dealers should carefully consider their tolerance for compliance risk in the event that an administration chooses an enforcement approach or if the firm is already in a precarious position.

Risk Tolerance

Because the SEC and FINRA have not issued guidance on the question of whether generative AI prompts and responses are communications subject to Rules 17a-3 and 17a-4, and they are unlikely to provide a dispositive answer other than the statement that the rules are technologically neutral, firms must make a risk decision on how to treat the data. Usually, each broker-dealer’s risk management program has buckets for risk acceptance, including a bucket for compliance or regulatory risk. Typically, firms adopt a low tolerance for compliance risk. Those firms should carefully consider whether their staff, including management, is using generative AI on unapproved and unmonitored providers for business purposes. Firms with a low compliance risk tolerance should consider a policy restricting generative AI use to approved providers, where records of prompts and responses are retained and the activity is supervised.

1. 17 C.F.R. § 240.17a-4(b)(4) (2026). ↑

2. See SEC v. Arete Wealth Mgmt., LLC, No. 25 C 616, slip op. at 27–32 (N.D. Ill. Feb. 26, 2026). Arete argued unsuccessfully that text messages concerning outside sales activities and compliance matters were not required to be retained because “business as such” was unconstitutionally vague, violating due process, and that some texts did not relate to the firm’s securities business. ↑

3. For example, OpenAI’s Terms of Use state, “We may use Content to provide, maintain, develop, and improve our Services, comply with applicable law, enforce our terms and policies, and keep our Services safe.” Terms of Use, OpenAI (Jan. 1, 2026). ↑

4. See United States v. Heppner, No. 1:25-cr-00503-JSR, 2026 WL 436479 (S.D.N.Y. Feb. 17, 2026) (finding that sourcing documents from a generative AI system invalidated a claim that the documents were privileged, even if the documents were subsequently provided to counsel). ↑

5. FINRA Rule 2210 Frequently Asked Questions D.8, FINRA (Dec. 22, 2025). ↑

6. Id. at FAQ B.4. ↑

7. Electronic Recordkeeping Requirements for Broker-Dealers, Security-Based Swap Dealers, and Major Security-Based Swap Participants, 87 Fed. Reg. 66412, 66428 (Nov. 3, 2022) (“One of the goals of this rulemaking is to make Rules 17a–4 and 18a–6 more technology neutral.”). ↑

8. SEC Charges 16 Wall Street Firms with Widespread Recordkeeping Failures, Sec. & Exch. Comm’n (Sept. 27, 2022). ↑

9. A Catalyst: Statement on Qatalyst Partners LP, Hester M. Peirce and Mark T. Uyeda, Comm’rs, Sec. & Exch. Comm’n (Sept. 24, 2024). ↑

The U.S. Court of Appeals for the Second Circuit decided four arbitration cases in 2025, each addressing a different piece of the process: agreement formation, scope, waiver, and judicial oversight of ongoing arbitration proceedings. Individually, they generated the usual case-specific commentary. But read together, they tell a bigger story: The Second Circuit is done treating arbitration agreements as a special category of contract that gets the benefit of the doubt. Going forward, they will be evaluated like any other agreement.

You Need Real Consent

Start with how agreements get made. In Davitashvili v. Grubhub Inc.,^[1] the court examined Grubhub’s checkout page and arbitration clause through a consumer-protection lens. In particular, the court focused on where the hyperlink to the arbitration provision sat, how cluttered the screen was, and whether the prompt was clear enough to put a reasonable user on notice. Under the circumstances presented, the clause survived—but barely. Judge Pérez wrote separately to warn that the page was “likely the high-water mark” for enforceability. In other words, a slightly more jumbled checkout screen likely would have defeated the clause altogether.

In Sudakow v. CleanChoice Energy, Inc.,^[2] the Second Circuit went further. In this decision, a company mailed an arbitration clause in a “Welcome Package” weeks after the customer had already signed an enrollment agreement, and nothing flagged the new terms. Simply receiving the Welcome Package and continuing to pay for electricity were not enough, the court held, to establish assent to arbitration.

The bottom line from both cases is that if you want an arbitration clause to survive a challenge in the Second Circuit, you need genuine assent at the point of contracting because constructive notice buried in later materials is not enough.

Broad Language Won’t Cover Every Claim

Davitashvili also addresses what arbitration clauses actually cover. Grubhub’s clause was extremely broad, covering disputes that “in any way relate to your use of the Platform.”^[3] Nonetheless, the court held that antitrust claims about anti-price-competition provisions between Grubhub and restaurants lacked a sufficient connection to an individual customer ordering dinner through the Grubhub app. Judge Sullivan dissented, arguing that a more lenient “meaningful nexus” standard should apply, while the majority required a more substantial link. The practical upshot is that where statutory claims (for example, antitrust, consumer protection, or securities claims) are conceptually distinct from the transaction governed by the general terms of service, you should not assume that a broadly worded arbitration clause will sweep them in.

Pick Your Forum Early or Lose It

Doyle v. UBS Financial Services, Inc.^[4] changed the Second Circuit’s approach to waiver of the right to arbitrate. Previously, a party could often avoid the waiver as long as the other side was not prejudiced by delay in seeking arbitration. Doyle removed that requirement, following the Supreme Court’s 2022 decision in Morgan v. Sundance, Inc.,^[5] which stripped courts of the ability to require a showing of prejudice before finding waiver. Now, the waiver turns on whether a party acted inconsistently with the right to compel arbitration. In Doyle, arbitration was waived where the defendant joined a motion to dismiss, waited to see how it played out, and only then moved to compel arbitration after an adverse decision. The days of “testing the waters” in court before falling back on arbitration appear to be over.

Once You’re In, You’re In

Frazier v. X Corp.^[6] addresses what happens after arbitration has started. In a mass-arbitration dispute involving former Twitter employees, the district court ordered the company to pay arbitration fees as assessed by JAMS. The Second Circuit reversed, holding that fee allocation is a procedural question for the arbitral forum to address; it was not a “failure, neglect, or refusal to arbitrate” that courts may remedy under the Federal Arbitration Act. The decision aligns the Second Circuit with the Third, Fifth, Ninth, and Eleventh Circuits. Because companies facing thousands of individual arbitration demands can be forced to pay millions of dollars in filing and administrative fees before the merits are even considered, court-ordered fee payments have become a central pressure point in mass-arbitration campaigns—and that lever has now been limited.

What It All Adds Up To

The pattern is clear. At every stage (formation, scope, waiver, and judicial oversight), the Second Circuit is applying the same basic principle: an arbitration agreement is a contract, and it gets treated like one. You need real consent. Your clause covers what it says it covers, not everything you wish it covered. You have to commit to arbitration early or risk losing it. And once the arbitration is underway, courts generally will not step in to manage the proceeding.

The immediate implication is a shift in leverage. Consumers and employees gain more room to challenge arbitration at the front end (especially where assent is imperfect or the clause is presented late), and plaintiffs with statutory claims may more often keep at least some theories in court when the dispute is only loosely connected to the underlying transaction. Businesses, by contrast, lose some of the “margin for error” that historically made arbitration a reliable default—and they lose the ability to wait and see how a case develops before invoking the clause.

None of this means that the Second Circuit is hostile to arbitration. Properly implemented, arbitration remains a powerful tool. But these recent 2025 decisions clarify who is likely to win and who is likely to lose. The practical takeaway is straightforward: make assent unmistakable; draft scope with precision; move to compel early; and expect procedural fights to be decided in arbitration, not federal court.

1. Davitashvili v. Grubhub Inc., 131 F.4th 109 (2d Cir. 2025). ↑

2. Sudakow v. CleanChoice Energy, Inc., 153 F.4th 280 (2d Cir. 2025). ↑

3. Davitashvili, 131 F.4th at 119. ↑

4. Doyle v. UBS Fin. Servs., Inc., 144 F.4th 122 (2d Cir. 2025). ↑

5. Morgan v. Sundance, Inc., 596 U.S. 411 (2022). ↑

6. Frazier v. X Corp., 155 F.4th 87 (2d Cir. 2025). ↑

When President Trump signed the Guiding and Establishing National Innovation for U.S. Stablecoins Act (the “GENIUS Act”) into law on July 18, 2025, the reaction across the digital asset industry was almost uniformly positive. At long last, the United States had delivered what the industry had been requesting for years: a comprehensive regulatory framework for dollar-backed stablecoins. Industry leaders and market participants welcomed the legislation and the clarity that it would offer the crypto industry.

On April 8, 2026, the U.S. Department of the Treasury’s Financial Crimes Enforcement Network (“FinCEN”) and the Office of Foreign Assets Control (“OFAC”) announced a joint notice of proposed rulemaking (the “Proposed Rule”) that would operationalize the GENIUS Act’s mandates. The Proposed Rule would classify permitted payment stablecoin issuers (“PPSIs”) as a new category of financial institution under the Bank Secrecy Act (“BSA”), separate from money services businesses (“MSBs”), and impose a comprehensive suite of anti-money laundering (“AML”), countering the financing of terrorism (“CFT”), and sanctions compliance obligations on PPSIs. Notably, the Proposed Rule would also be the first federal regulation to codify what constitutes an “effective” OFAC sanctions compliance program, attaching penalties for PPSIs that fail to maintain required program elements—even absent underlying sanctions violations.

The FinCEN/OFAC Proposed Rule does not exist in isolation. As described below, it is part of a broader, multi-agency implementation effort under the GENIUS Act. Stablecoin issuers should be monitoring these rulemakings in parallel, particularly those subject to overlapping federal and state jurisdiction. Separately, FinCEN has also issued a proposed rule that would overhaul AML/CFT program requirements more broadly for a wider range of financial institutions, which shares structural similarities with the PPSI-specific rule and may result in overlapping obligations for companies subject to both.

The Proposed Rule has significant implications for current and prospective stablecoin issuers, their parent institutions, and the broader digital asset ecosystem. Comments are due June 9, 2026, and final regulations are expected by July 18, 2026, with the effective date scheduled for January 2027.

This article discusses the Proposed Rule’s key elements, beginning with a primer on the GENIUS Act framework.

The GENIUS Act

Understanding the Proposed Rule requires a brief detour through the statute it implements. The GENIUS Act is the first U.S. federal regulatory framework for payment stablecoins—defined under the statute as a digital asset that is used or designed for use as a means of payment or settlement, where the issuer is obligated to redeem it for a fixed monetary value and represents that it will maintain a stable value relative to that amount (in practice, one U.S. dollar). The statute provides the answers to questions regulators had been sidestepping for years: What exactly is a stablecoin, who is permitted to issue one, and how are they regulated?

The statute answers the last question with a licensing framework. Only “permitted payment stablecoin issuers,” or PPSIs, may issue payment stablecoins in the United States. There are three pathways to PPSI status: a PPSI may be (1) a subsidiary of an insured depository institution approved by its primary federal banking regulator to issue stablecoins; (2) a federally qualified payment stablecoin issuer, including nonbank entities chartered by the Office of the Comptroller of the Currency (“OCC”); or (3) a state-qualified payment stablecoin issuer approved by a state regulator meeting federal standards. Knowingly issuing payment stablecoins without PPSI authorization can result in fines of up to $1 million per violation and imprisonment for up to five years under the GENIUS Act.

The GENIUS Act also constrains what PPSIs can do. Their authorized activities are limited to issuing and redeeming payment stablecoins, managing reserve assets (which must back outstanding stablecoins on at least a 1:1 basis), and providing custodial services.

Two additional features of the GENIUS Act are worth flagging. First, the statute amends the federal securities laws and the Commodity Exchange Act to provide that qualifying payment stablecoins issued by PPSIs are excluded from the definitions of both “security” and “commodity.” This is a significant jurisdictional carve-out: The Securities and Exchange Commission and Commodity Futures Trading Commission do not regulate payment stablecoins under the GENIUS Act. Instead, oversight falls to banking regulators—the OCC, the Federal Deposit Insurance Corporation (“FDIC”), the Board of Governors of the Federal Reserve System, the National Credit Union Administration (“NCUA”), and state banking regulators—along with FinCEN and OFAC for illicit finance and sanctions. Second, the GENIUS Act contains bankruptcy protections for stablecoin holders, including a priority claim senior to all other creditors and an exclusion of reserve assets from the bankruptcy estate.

The GENIUS Act requires that the Department of the Treasury, including FinCEN and OFAC, as well as the federal banking agencies, promulgate rules to implement the GENIUS Act. The Proposed Rule addressed in this article is part of this multi-agency effort, and its implications for stablecoin issuers’ compliance infrastructure are among the most significant of the proposed rules to date. Since the GENIUS Act’s enactment, the designated regulatory agencies have moved quickly to implement the statute. The OCC published a proposed rule in February 2026 establishing a comprehensive prudential, operational, and supervisory framework for PPSIs under its jurisdiction. The FDIC followed in April 2026 with a proposed rule addressing PPSI subsidiaries of FDIC-supervised insured depository institutions. The NCUA has published proposed rules as well. The Treasury Department established principles determining whether state-level regulatory regimes are “substantially similar” to the federal framework. In addition, FinCEN issued a separate proposed rule in the same week as the Proposed Rule that would overhaul AML/CFT program requirements for a wider range of financial institutions, including by encouraging financial institutions to consider the use of digital identity, blockchain analytics, and generative artificial intelligence to prevent financial crime.

A New Category of Financial Institution

The Proposed Rule’s most consequential result is the creation of a new, standalone category of financial institution under the BSA. Many stablecoin issuers have historically been subject to BSA obligations as money transmitters, a subcategory of MSBs, under FinCEN rules. Under the Proposed Rule, PPSIs would no longer be classified as MSBs. Instead, they would be regulated under a separate framework with some obligations mirroring those applicable to banks.

This is not a cosmetic change, as PPSIs will become subject to a range of requirements that MSBs are not, including enhanced due diligence for correspondent and private banking accounts and compliance with special measures when foreign financial institutions or transactions are deemed to be of primary money laundering concern. In practice, this means that if FinCEN designates a foreign jurisdiction or institution as being of primary money laundering concern—as the Trump administration has recently done with certain Mexican financial institutions in its push to address illicit drug trafficking—PPSIs with exposure to those jurisdictions or institutions would need to comply with any restrictions or enhanced due diligence requirements imposed.

The reclassification also reflects FinCEN’s stated assessment in the Proposed Rule that the economic functions PPSIs perform more closely resemble those of banks than those of traditional money transmitters. In short, PPSIs issue a widely used medium of exchange, maintain reserve assets, and facilitate payments—functions that in key respects more closely resemble banking than traditional money transmission. The Proposed Rule relies on FinCEN’s authority under the BSA to designate PPSIs as businesses engaged in activities “similar to, related to, or a substitute for” the activities of enumerated financial institutions.

Primary and Secondary Markets

The Proposed Rule divides the stablecoin ecosystem into primary and secondary markets, with different compliance obligations in each.

The “primary market” covers transactions where a PPSI interacts directly with a user or holder of a payment stablecoin. Issuance, redemption, and direct transfers are primary market activity. In the primary market, the full suite of AML/CFT obligations applies: customer identification, ongoing due diligence, suspicious activity monitoring, and the other requirements discussed below.

The “secondary market” covers transactions where the PPSI has no direct interaction with the transacting parties. Value moves between users through the PPSI’s smart contract without any direct customer relationship. This is the peer-to-peer transfer environment, the on-chain world where stablecoins circulate between wallets, exchanges, and DeFi protocols without the issuer’s active involvement in any given transaction.

The Proposed Rule does not require PPSIs to file suspicious activity reports or conduct customer due diligence on secondary market activity. However, it does require PPSIs to maintain the technical ability to block, freeze, and reject transactions across their entire network, including the secondary market. This is where much of the real compliance cost resides. If OFAC designates a wallet address, the issuer must be able to prevent that address from transacting with its stablecoin, even though the issuer has no customer relationship with the owner of the wallet. In other words, PPSIs would bear no affirmative obligation to monitor or report on secondary market transactions, but they would bear a full obligation to intervene in those same transactions when directed by law, regulation, or sanctions designation. That tension—between passive observation and active enforcement capability—is where much of the industry commentary is likely to focus.

AML/CFT Program Requirements

In the primary market, the Proposed Rule would explicitly require PPSIs to conduct ongoing customer due diligence, including complying with “know your customer” requirements and monitoring accounts for suspicious transactions. Many stablecoin issuers that currently operate as MSBs conduct some form of customer identification, but the Proposed Rule would implement new expectations.

The Proposed Rule would require PPSIs to file suspicious activity reports (“SARs”) with FinCEN for any primary market transaction involving $5,000 or more, adopting the same dollar threshold that currently applies to banks. This is a notable departure from the MSB framework, where the SAR filing threshold for money transmitters is $2,000. The higher threshold may reduce volume, but the underlying obligation to detect and report suspicious activity across the full range of primary market transactions is no less demanding.

PPSIs would be required to retain records on transfers of $3,000 or more and to share specified information with other financial institutions involved in fund transfers. This recordkeeping threshold mirrors the existing requirement for financial institutions under the BSA’s funds transfer rules. The information-sharing obligation is designed to create an auditable trail across institutions, enabling regulators and law enforcement to reconstruct the flow of funds when investigating potential illicit activity.

The Proposed Rule would require PPSIs to collect beneficial ownership information for business customers, an obligation that does not currently apply to MSBs and that represents one of the more operationally significant new requirements. Issuers will need to build infrastructure for identifying and verifying the natural persons who ultimately own or control their business counterparties. For issuers whose customer base includes institutional participants, exchanges, or other entities, this requirement may necessitate new data collection workflows.

The Proposed Rule would extend to PPSIs two obligations drawn directly from the USA PATRIOT Act that apply to banks: enhanced due diligence programs for correspondent and private banking accounts, and the obligation to respond to government information-sharing requests. In the PPSI context, correspondent account due diligence would require issuers to assess and manage the risks posed by relationships with foreign financial institutions that hold accounts with the PPSI or use the PPSI’s payment infrastructure, while private banking account due diligence would impose heightened scrutiny on high-value individual accounts. For issuers that serve institutional customers, foreign exchanges, or other entities that function as intermediaries, these requirements could necessitate substantial new compliance processes for onboarding and ongoing monitoring of those relationships.

PPSIs would become subject to both the mandatory and voluntary information-sharing provisions of Section 314 of the USA PATRIOT Act. Under the mandatory provision (Section 314(a)), FinCEN and law enforcement agencies can query PPSIs about whether they maintain accounts or have conducted transactions for specified individuals or entities. Under the voluntary provision (Section 314(b)), PPSIs would be permitted to share information with one another and with other financial institutions for purposes of identifying and reporting potential money laundering or terrorist financing activity, with a safe harbor from liability for such sharing.

The illicit finance data motivating these requirements is substantial. According to the Proposed Rule, between January 2015 and November 2025, FinCEN received approximately 55,000 suspicious activity reports referencing specific stablecoins, and OFAC received approximately 5,800 blocked property reports and 3,000 rejected transaction reports referencing stablecoins. The Proposed Rule cites the use of stablecoins in money laundering chains, North Korean cyber theft, sanctions evasion networks, fentanyl precursor procurement, and terrorist financing as the specific risks driving the rulemaking.

The First Codified Sanctions Compliance Program Requirements

Perhaps the most significant regulatory development in the Proposed Rule is OFAC’s decision to codify, for the first time in its regulations, what constitutes an “effective” sanctions compliance program and penalties for failing to maintain one.

All U.S. persons are already required to comply with OFAC-administered sanctions, including stablecoin issuers. But until now, OFAC has never required any category of U.S. person to maintain a formal compliance program with potential penalties for failure to do so. OFAC’s longstanding “Framework for OFAC Compliance Commitments,” published in 2019, laid out a five-element structure for an effective sanctions program: management commitment, risk assessment, internal controls, testing and auditing, and training. Additionally, OFAC’s 2021 “Sanctions Compliance Guidance for the Virtual Currency Industry” specifically addressed sanctions compliance best practices for the virtual currency industry. However, both materials represented regulatory guidance, not a binding legal obligation.

The GENIUS Act changes this. The statute directs that PPSIs maintain an “effective economic sanctions compliance program . . . consistent with [f]ederal law,” and the Proposed Rule operationalizes that directive by incorporating OFAC’s five-element framework into binding regulation. A PPSI’s sanctions compliance program must include: management commitment from senior leadership; a risk assessment process that identifies sanctions risks specific to the issuer’s products, customers, and geographic exposure; risk-based internal controls to identify, block, and reject transactions that may violate sanctions; independent testing and auditing of the program’s effectiveness; and training for all relevant personnel.

This represents a meaningful paradigm shift. For years, OFAC’s enforcement posture relied on the implicit message that the voluntary framework was, in practical terms, non-optional. OFAC would consider a company’s compliance infrastructure, or lack thereof, when determining potential enforcement or penalties for sanctions violations. But the gap between favorable treatment for maintaining a good program and a legal requirement to have one, backed by penalties for failure to maintain it, is significant. Practitioners should expect it to serve as a template for other regulated industries. For this reason, we anticipate industry commentary to focus on this net-new requirement—especially as it is in direct tension with Treasury’s broader effort in a separate proposed rule to focus enforcement actions on outcomes rather than “foot fault” program deficiencies.

Block, Freeze, Reject, and Burn

The Proposed Rule’s most operationally demanding requirement is arguably its mandate that PPSIs maintain the technical capability to block, freeze, and reject impermissible transactions, as well as comply with lawful orders to seize, freeze, burn, or prevent the transfer of payment stablecoins.

These requirements extend to both primary and secondary market activity. In the primary market, where the issuer has a direct relationship with the customer, the block-and-freeze architecture maps reasonably well onto existing bank compliance workflows. In the secondary market, however, the requirement has no direct parallel for other BSA-regulated financial institutions. Banks are not generally required to control transactions between third parties with whom they have no customer relationship. But a stablecoin issuer, whose product circulates freely on public blockchains, would be required to maintain the infrastructure to intervene in peer-to-peer transfers if required by a lawful order or sanctions designation.

In practice, this means issuers will need on-chain compliance infrastructure that monitors activity with which they have no direct commercial relationship. That means transaction screening systems, smart-contract-level controls, and the engineering capacity to execute freezes and burns in response to regulatory directives. The engineering investment required to build and maintain these capabilities is substantial. The inclusion of “burn”—meaning the permanent and irreversible destruction of tokens, effectively removing them from circulation with no possibility of recovery by the holder—is notable. Unlike a freeze, which temporarily restricts access, or a block, which prevents a specific transaction, a burn eliminates the tokens entirely. While some issuers have built burn capabilities into their smart contracts, codifying this as a regulatory requirement raises novel questions about property rights, due process, and the limits of issuer control over circulating tokens.

This Is a Proposed Rule

Currently, this remains a notice of proposed rulemaking. The rule is not final and is subject to revision based on public comment. Those revisions could be meaningful.

The comment period is open, and comments are due June 9, 2026. Final regulations are expected by July 18, 2026, consistent with the GENIUS Act’s one-year implementation deadline, with enforcement beginning no later than January 2027.

The sixty-day comment period presents a genuine opportunity. Several aspects of the Proposed Rule invite industry input. The “burn” requirement and secondary market compliance obligations, particularly the scope of the block-and-freeze mandate for peer-to-peer transactions, are likely to generate significant comment. The codification of a burn requirement—authorizing the permanent destruction of tokens—raises novel questions about property rights and due process that are also likely to draw substantial feedback. The interaction between the new PPSI category and existing MSB registrations raises transitional questions that FinCEN has acknowledged but not fully resolved. The OFAC sanctions program requirements, while largely tracking the voluntary framework, introduce new specificity around what “effective” means in practice, and industry participants may have views on calibration. The SAR and recordkeeping thresholds, inherited from the bank framework, may warrant stablecoin-specific adjustments.

The final rule could differ meaningfully from the proposal based on feedback received. The rulemaking record will shape not only the contours of the final regulation but also the interpretive guidance and examination expectations that follow it. For issuers, exchanges, custodians, compliance technology providers, and the institutions that bank stablecoins, this is the moment to shape the framework that will govern the U.S. stablecoin market for the foreseeable future.

Nevertheless, payment stablecoin issuers should not wait for the final rule to begin preparing. The scope and detail of the Proposed Rule make clear that significant compliance changes are coming, and issuers should be evaluating their programs now to ensure they are positioned to meet these new obligations when enforcement begins.

In April 2026, the New York Times profiled the founder of Medvi, a new AI-enabled telehealth company selling weight-loss drugs. So far, Medvi is on track to do $1.8 billion in revenue this year with just two employees. Like every fast-growing company operating in a highly regulated industry, Medvi has significant and growing legal exposure. And like most companies at that stage, its founder is not looking to become the quarterback of a complex legal function.

Every company needs a legal partner who understands its business and takes ownership of that exposure as it grows. Unfortunately, the legal market has never built that.

This market gap is not AI-driven. Companies have always sought to focus on their core business. As legal needs grow and regulatory complexity increases, companies have been forced to build internal legal expertise—not because they wanted to, but because the market never offered a better option.

The Model Was Never Built for Clients

The prevailing legal model is the billable hour, which doesn’t just misalign incentives: It structurally prevents firms from delivering what clients want. Compensation structures reward hours, not outcomes. Origination credit belongs to someone who may no longer provide any value to the client. Referring a client to a colleague inside the firm often carries more personal financial risk than reward for the lawyer doing the referring. The result is that firms with dozens of practice areas may serve the same client as a series of disconnected engagements, while the client bears the cost of coordinating across teams that have no formal incentive to work together.

The model doesn’t just fail clients; it creates challenges for lawyers. Firms expect partners to serve as subject matter experts, business developers, relationship managers, mentors, and technology adopters while rewarding billing activity above nearly everything else. The result is a structure that makes client-centered collaboration difficult. Good lawyers know that this dynamic creates a fundamentally flawed structure and is part of why capable attorneys have been leaving Big Law in meaningful numbers. Not because they don’t want to do excellent work, but because the model and resulting culture prevent them from doing it in a way that feels aligned with real client value.

Corporate and institutional clients are growing their internal legal departments and bringing more work in-house because there is no compelling alternative. Thomson Reuters’ 2026 State of the Corporate Law Department Report confirms what the in-house trend has obscured: Companies do not want to be in the business of managing legal services. Despite years of insourcing, nearly half of general counsel cite resource and staffing constraints as their top barrier to delivering value, and only 17 percent of C-suite executives view their legal department as a significant organizational contributor.

General counsel are expected to be strategic legal partners to the business while simultaneously managing outside counsel, coordinating legal work across multiple specialties, and identifying risks before they become problems. In many organizations, they are carrying responsibility for a function that remains fragmented by design.

AI Has Not Changed the Fundamental Dynamic

Generative AI has not created a crisis in the legal market; it has revealed one that was already there. And it is arriving at exactly the moment when legal complexity is accelerating.

Law firms are responding by investing in AI to protect and extend the existing model. They are offering faster research, smarter document review, and more efficient billing, all at higher rates. The goal is for their lawyers to do more of the same work in about the same way with the same economics intact. Separately, AI-native companies are going directly to clients, bypassing law firms entirely for commoditized work. Sequoia Capital has mapped $20–25 billion in legal transactional work, contract drafting, nondisclosure agreements, and regulatory filings—all work that AI can now deliver directly to the client.

Despite their differences, both approaches leave the same responsibility in the same place: with the client. Whether the work is performed by outside counsel or a tech platform, the company is still expected to identify legal issues, determine what expertise is required, coordinate solutions across disciplines, and manage the outcome.

The Model That Has to Be Built from Scratch

Consider a fast-growing healthcare company operating across multiple states. Under the current model, the company may separately retain employment counsel, regulatory counsel, privacy counsel, litigation counsel, and commercial contract counsel, while its internal legal department coordinates the risks and communication between them. The client is responsible for connecting issues across the business, identifying emerging exposure, and determining when legal intervention is needed.

A different model would reverse that burden. A law firm built specifically around healthcare companies would proactively monitor regulatory developments, coordinate legal strategy across disciplines, identify patterns before they become disputes, and manage recurring operational legal needs through integrated workflows and technology-enabled systems. Instead of paying multiple disconnected providers to react to problems after they emerge, clients would operate within a coordinated system designed to identify risk earlier, reduce duplication, create institutional knowledge, accelerate response times, and lower the overall cost of managing legal exposure.

A full-service law firm built around a specific industry vertical and using AI where it makes the most sense starts from a blank page. The model is designed around a complete understanding of what clients in a specific industry actually need across their full legal function, then built up from there using best practices from business and professional services firms. That is a different exercise than redlining a traditional law firm model. Every element of the model, from compensation to technology to service delivery, is designed around client outcomes. Lawyers focus on what only lawyers can do. Dedicated professionals manage workflow, track obligations, and anticipate what clients will need next. Generative AI is deployed where it drives the most value for the client, eliminating the friction that inflates cost, slows delivery, and prevents proactivity in legacy models. The firm’s team sees the full picture and takes the work off the client’s plate entirely.

Traditional law firm remuneration makes this structurally impossible to build within most existing firms—not because the desire doesn’t exist, but because the incentives prevent it. Generative AI is making the economics of the traditional model less and less viable, and creating room for a model that law firms never had a financial incentive to build.

Building a legal services operation that actually delivers what full-service law firms have always promised is what corporate clients want, but what the market has never produced. The ability to build it exists and that makes this a race. The people who move first will be the ones who finally give the market an option that it has been asking for.

Beyond a lawyer’s professional responsibility to provide services to those who cannot afford to secure legal representation, pro bono work is emerging as a powerful vehicle for strategic, lasting, and transformative change in communities.

At the ABA Business Law Section 2026 Spring Meeting held in Atlanta, Georgia, the ABA Business Law Section Pro Bono Committee hosted the roundtable “Investing in High-Impact Pro Bono Projects,” discussing how legal professionals, including transactional attorneys, empower businesses through pro bono legal assistance and leverage innovative models to scale impact. The roundtable discussion, moderated by Sandrine Siewe, ABA Business Law Fellow, featured insights from Radha Sathe Manthe, Pro Bono Deputy at King & Spalding LLP; Kerry-ann Archer, Senior Counsel at Wells Fargo Legal Department; Kate M. Gaffney, Pro Bono Director at Atlanta Legal Aid Society; and Alina Lee, Founding Partner, Aspire Law.

High-Impact Pro Bono Opportunities for Transactional Attorneys in Atlanta

When explaining how their organizations identify high-impact pro bono opportunities, Manthe and Lee emphasized that they ground their pro bono assistance in what communities need through strong collaboration with their local partners, such as Georgia Justice Project, Atlanta Legal Aid Society, and Pro Bono Partnership of Atlanta.

At Aspire Law, as Lee explained, representing or assisting pro bono clients with legal matters gives lawyers a sense of connectedness and purpose, assuring them that their efforts matter and they make a difference in their community. A crucial step Aspire Law takes in identifying impactful pro bono opportunities is consulting legal aid nonprofits about the practice areas with the most pressing needs, drawing on their expertise on the ground. Some of the firm’s pro bono services include creating and sharing contract templates with Pro Bono Partnership of Atlanta and providing a training program for nonprofit board members that outlines best practices in board governance and fiduciary duties. To date, this training has been rolled out to over a hundred nonprofits across Georgia.

At King & Spalding, pro bono work is informed by community needs through several partnerships, including with Georgia Justice Project and Atlanta Legal Aid Society. Many of their transactional lawyers enjoy assisting small businesses and nonprofit organizations with their legal needs. Entrepreneurs often lack access to legal advice early on, and assistance with issues such as entity formation, contracts, and governance can be critical to helping those organizations launch and grow. The firm’s pro bono work also includes immigration matters, which, Manthe emphasized, can be transactional in nature and do not always involve court appearances. Attorneys may assist with drafting asylum applications, preparing affidavits, and gathering relevant evidence. Although immigration work may sound intimidating to some transactional attorneys, their attention to detail is a useful skill set for immigration applications.

Partnerships Between Law Firms, Corporations, and Legal Service Providers

Community-informed pro bono services require trust and sustained relationships between law firms, corporations, and legal service providers. To ensure a successful and sustainable collaboration, organizations should be intentional about the values that guide their relationship. At Atlanta Legal Aid Society, client needs and mission alignment are central to partner selection. Gaffney noted that Atlanta Legal Aid Society is grateful for partners who are flexible in meeting evolving client needs rather than defining those needs themselves. Actions that make strong partnerships include firms offering to staff all cases on a clinic date, piloting a new project, offering candid feedback and allowing for ongoing improvements, reassigning cases internally when an associate leaves, and developing project mentors within firms.

In legal departments, partnerships with local partners may have a different structure. There is a growing trend toward the development of joint pro bono ventures between in-house and outside counsel that allow each party to the relationship to leverage their particular strengths and skills. For instance, Wells Fargo collaborates with law firms and legal service providers through a variety of channels, including the Charlotte Triage Pro Bono Partnership. This initiative was launched in 2018, bringing together law firms and corporations to focus on areas of greatest need identified by local legal aid organizations. Through this partnership, Wells Fargo in-house counsel assist indigent clients with a variety of matters ranging from advising consumers and small businesses to helping tenants across North Carolina with housing issues.^[1] For instance, Archer has assisted small and emerging businesses on various matters. She has also had the opportunity to work on housing-related matters, including eviction defense work to prevent homelessness. These matters often involve urgent circumstances for families, and even limited legal intervention can make a significant difference in stabilizing housing and preventing displacement.

Such strategic partnerships help coordinate participation of legal professionals and ultimately, build civic infrastructure needed for improved access to justice. Indicators of success beyond case outcomes include repeat collaborations, program renewals, and increased participation from legal professionals.

Innovative Models of Pro Bono Support and the Future of Access to Justice

As the demand for pro bono services continues to outpace available resources, the panelists highlighted some innovative approaches and technologies their organizations and clients are adopting or considering to address longstanding challenges in the pro bono sector, including resource constraints and access barriers. With a view to optimizing available resources, Atlanta Legal Aid Society may offer very defined litigation opportunities with detailed training and mentorship for transactional lawyers interested in experience outside their area, such as temporary protective order matters, name change cases, or eviction clinics. To help scale up pro bono work, other organizations are adopting tools powered by artificial intelligence when providing legal services, while taking into account ethical risks related to emerging technologies. For instance, one of Lee’s clients, the legal department of an Atlanta-based company, is actively training an AI agent to answer the most common legal questions for pro bono matters, making it easier and faster for their attorneys to provide pro bono legal support. The company also plans to make the AI agent available to pro bono organizations after adequate testing for greater impact.

The panelists’ insights reflect how firms can thoughtfully invest their business law resources and transactional talent in high-impact pro bono work. The outcomes of such an investment are practical and sustainable: a nonprofit that can now legally raise funds, a small-business owner who fully understands their contractual obligations, and a community organization with governance structures and systems that outlast its founder’s departure.

ABA National Public Service Award

Each year, the ABA Business Law Section presents the National Public Service Award as part of its efforts to recognize significant pro bono legal contributions in a business context. Following the roundtable, the ABA Business Law Section Pro Bono Committee presented the 2026 National Public Service Award to Fenwick & West LLP and Justin B. Finn, for their outstanding commitment to providing legal services to individuals and entities that could not otherwise afford them. Amid rapid policy changes, natural disasters, and economic uncertainty, Fenwick provided 24,256 hours of pro bono legal services, raised almost $500,000 for organizations addressing food insecurity and disaster relief, and hosted more than 120 pro bono clinics and high-impact programs serving those most affected. Finn is widely recognized for his leadership in advancing bipartisan civic engagement and public service through innovative legal and community initiatives that bridge societal divides while promoting collaboration, dialogue, and democratic participation.

This article reflects insights from the 2026 ABA Pro Bono Happy Hour Roundtable held in Atlanta, Georgia, and organized by Jayme Cassidy and Brenda Barrett, Co-Chairs of the ABA Business Law Section Pro Bono Committee.

1. Charlotte Triage was launched by pro bono leaders from Bank of America, McGuireWoods, Duke Energy, Moore & Van Allen, Wells Fargo and Husqvarna. ↑

This article is Part XI of the Musings on Contracts series by Glenn D. West, which explores the unique contract law issues the author has been contemplating, some focused on the specifics of M&A practice, and some just random.

What does a hypothetical, annoyingly intrusive, know-it-all person, who observes your and your counterparty’s negotiations and makes unrequested and purportedly authoritative suggestions about specific provisions to include in your contract, have to do with the implied covenant of good faith and fair dealing in Delaware? In other words, what does Lord Justice MacKinnon’s “officious bystander” analogy from the 1939 English case of Shirlaw v. Southern Foundries (1926) Ltd.^[1] have to do with Delaware’s particular approach to the implied covenant? The answer was provided by Vice Chancellor Laster in a recent Delaware Court of Chancery decision, Guilbeau v. Footprint International Holdco, Inc.^[2]

A Basic Primer on Delaware’s Approach to the Implied Covenant

As everyone should know, Delaware, like many other (but not all)^[3] states, purports to recognize that “[e]very contract imposes upon each party a duty of good faith and fair dealing in its performance and enforcement.”^[4] But however lofty and far-reaching that concept may sound, and no matter how often it may be invoked by a plaintiff as “an equitable remedy for rebalancing the economic interests [of the contracting parties] after events that could have been anticipated but were not,”^[5] the Delaware Supreme Court has declared that “the covenant functions as a limited ‘gap-filler’: it enforces the parties’ reasonable expectations in circumstances that they could not foresee and did not address in their written agreement, but it may not be used to rewrite or contradict express terms.”^[6] Thus, despite its name, the implied covenant of good faith and fair dealing, as applied in Delaware and perhaps elsewhere, does not appear to create an actual implied obligation requiring contracting parties to perform their contract fairly and in good faith.

Indeed, the Delaware Supreme Court has “repeatedly emphasized that the implied covenant’s gap-filling power is a ‘limited and extraordinary remedy’; a ‘cautious enterprise’ that applies only where there is a true contractual gap about how to handle an unforeseen event.”^[7] It is “reserved for developments that could not have been anticipated, not developments that the parties simply failed to consider.”^[8]

There are two primary circumstances in which the implied covenant can apply in Delaware:

The first is when a contract allocates discretionary authority to one party over a central aspect of the contract. When the party exploits that discretion in a manner that defeats the “overarching purpose” of the bargain, courts may imply a requirement that such discretion be exercised reasonably and in good faith to ensure that the discretionary power is applied consistently with what reasonable parties would have agreed to at signing. . . .

The second use is . . . to address unforeseen developments—contingencies neither anticipated nor resolved by the contract—that threaten the parties’ bargained-for economic expectations. The law recognizes that “[n]o contract, regardless of how tightly or precisely drafted it may be, can wholly account for every possible contingency.” When such an unanticipated development arises, “the court has in its toolbox the implied covenant of good faith and fair dealing to fill in the spaces between the written words.”^[9]

The Claims in Footprint International

Footprint International involved a dispute over a “cram-down financing.”^[10] The plaintiffs were early investors in the company’s Class A preferred stock. In connection with their original investment, they entered into a Governance Agreement that was subsequently amended several times. Pursuant to the Governance Agreement, the majority of the Class A preferred stock “designated” one director of the company (“Class A Director”). In early versions of the Governance Agreement, the Class A preferred stockholders were granted a “liquidation preference equal to 1.4x of the purchase price plus the top spot in the liquidation distribution waterfall.”^[11] And importantly, the Governance Agreement “prohibited the Company from changing the Class A stock’s ‘rights, powers or preferences’ except with approval from a Board majority that included the affirmative vote of the Class A Director.”^[12]

When the company faced a severe liquidity crisis and potential insolvency, it entered into a series of financing transactions, the last of which (“Class F Financing”) “had a devastating effect on the Class A Stockholders.”^[13] The Governance Agreement was effectively amended so that it “eliminated all of the Class A protections[;] [i]t also eliminated the Class A Director.”^[14] But it appears that the Class A Director had voted in favor of these amendments.

The plaintiffs, who were the early-stage Class A stockholders, sued, claiming, among other things, that the Governance Agreement contained implied covenants that the “Class A Director was to act in the best interests of the Class A stockholders [and] . . . vote against any transactions that would harm the Class A stockholders.”^[15] They also argued that the Governance Agreement contained an implied covenant that effectively granted the Class A stockholders a veto over any amendment that would defeat their preferences and allow the issuance of other stock superior to the Class A stock.^[16] They further argued that the parties to the Governance Agreement “had an implied obligation to use their discretionary rights so as to protect and preserve the Class A stockholders’ rights.”^[17]

The defendants moved to dismiss these claims at the pleading stage. And Vice Chancellor Laster granted the defendants’ motion to dismiss, finding that “[n]one of the allegedly implied terms is reasonably conceivable.”^[18]

Vice Chancellor Laster’s Explanation of Delaware’s Jurisprudence on the Implied Covenant

In rejecting the plaintiffs’ claims on a motion to dismiss, Vice Chancellor Laster made several important observations regarding Delaware’s limited “gap-filling” approach to the implied covenant of good faith and fair dealing.

First, invoking the implied covenant involves a three-step process. The court must initially determine whether the contract contains a gap. Next, the court must determine whether that gap should be filled. Lastly, the court must determine which implied term should fill that gap.^[19]

Second, Vice Chancellor Laster made clear that the implied covenant can be invoked only after “the court has determined what the contract explicitly contemplates”^[20] because “it cannot be invoked where the contract itself expressly covers the subject at issue.”^[21] Only then can a determination be made as to whether a gap exists to be filled by the implied covenant.

Third, “[t]he implied covenant seeks to enforce the parties’ contractual bargain by implying only those terms that the parties would have agreed to during their original negotiations if they had thought to address them.”^[22]

Fourth, even when a contract appears to contain a contractual gap, there may be built-in contract-law default provisions that fill the gap, and the parties negotiating the contract were presumed to have made those provisions part of their contract because “[p]arties negotiate in the shadow of default principles of law.”^[23] And those default principles appear to include prior judicial precedent interpreting an agreement containing similar provisions.^[24]

For example, Delaware precedent holds that the term affiliates in a restrictive provision of a contract can be deemed to include not only affiliates existing on the contract date but also persons who thereafter become affiliates.^[25] Failing to state whether future affiliates are included does not create a gap to be filled. Delaware precedent fills that purported gap because by failing to exclude future affiliates, the drafters have included them.^[26] Thus, when using specific terms without qualifications or exceptions, you may be adopting prior judicial interpretations of those terms into your contract.^[27]

Fifth, Vice Chancellor Laster noted that the oft-repeated statement that the implied covenant can only be applied in situations where the parties “could not have anticipated” the situation that ultimately confronted the parties is not “literally true because the Delaware Supreme Court has recognized that the ‘parties occasionally have understandings or expectations that were so fundamental that they did not need to negotiate about those expectations.’“^[28] Furthermore, “[t]erms so obvious that both sides implicitly understood them are, necessarily, terms that could have been anticipated[;] [i]ndeed, they were both anticipated and known, yet the implied covenant can address them because they were so basic that no one would have thought to include them in the agreement.”^[29]

Lastly, applying the implied covenant to a party’s exercise of its discretion is not as far-reaching as it might seem. As other cases have made clear, even when applied to discretionary acts, the implied covenant is still only a gap filler—“if the scope of discretion is specified, there is no gap in the contract as to the scope of the discretion, and there is no reason for the Court to look to the implied covenant to determine how discretion should be exercised.”^[30] Moreover, Vice Chancellor Laster made clear that a party may exercise its discretion to advance its self-interest without violating the implied covenant of good faith and fair dealing. What it cannot do, however, is “wield a discretionary contractual right like a mafia gangster by using it to inflict harm on the counterparty unless the counterparty does what it wants.”^[31]

English Law Provides Helpful Guidance in Applying Delaware’s Gap-Filling Approach to the Implied Covenant

Although English law has not fully embraced the implied covenant of good faith and fair dealing,^[32] Vice Chancellor Laster noted that “English law has developed helpful answers that go beyond the current state of Delaware jurisprudence” in determining when contractual terms should be implied in a contract generally.^[33]

Vice Chancellor Laster then quoted from a “leading Privy Council judgment” as follows:

[F]or a term to be implied, the following conditions (which may overlap) must be satisfied: (1) it must be reasonable and equitable; (2) it must be necessary to give business efficacy to the contract, so that no term will be implied if the contract is effective without it; (3) it must be so obvious that “it goes without saying”; (4) it must be capable of clear expression; [and] (5) it must not contradict any express term of the contract.^[34]

Regarding the “issue of obviousness,” Vice Chancellor Laster next quoted Lord Justice MacKinnon’s well-known (at least to English lawyers) “officious bystander test”:

Prima facie that which in any contract is left to be implied and need not be expressed is something so obvious that it goes without saying; so that, if, while the parties were making their bargain, an officious bystander were to suggest some express provision for it in their agreement, they would testily suppress him with a common “Oh, of course!”^[35]

Vice Chancellor Laster then summarized the English approach to implying terms as follows:

Thus, under English law, “a term should not be implied into a detailed commercial contract merely because it appears fair or merely because one considers that the parties would have agreed [to] it if it had been suggested to them.” The term must also “be so obvious as to go without saying or to be necessary for business efficacy.” The additional requirements limit the court’s flexibility in deploying the implied covenant, but using more tractable concepts than Delaware law has yet deployed.^[36]

Vice Chancellor Laster then “borrowed” the English framework for implying terms generally to assess whether the plaintiffs’ asserted implied terms were appropriate under Delaware’s approach to the implied covenant of good faith and fair dealing. Indeed, he repeatedly asked in the opinion whether the parties to the Governance Agreement would have responded with a unanimous “Of course!” if an “officious bystander” observing their negotiations had suggested the specific terms that the plaintiffs claim were implied.^[37]

The answer in most cases was not “of course,“ but instead was “of course not.” But by asking the question, Vice Chancellor Laster was using this English framework to assist in doing what the Delaware Supreme Court had previously decreed that the implied covenant of good faith and fair dealing was “well suited” to do—i.e., “imply contractual terms that are so obvious . . . that the drafter would not have needed to include the conditions as express terms in the agreement.”^[38]

There Was No Implied Covenant That the Class A Director Was Only Tasked with Acting on Behalf of the Class A Stockholders

Delaware law makes clear that corporate directors “owe fiduciary duties to the entity and the entire body of stockholders generally rather than to individual stockholders or stockholder subgroups.”^[39] That requirement is a built-in default rule. Although the Governance Agreement did not specify the Class A Director’s duties, there was no gap in the agreement for the implied covenant to fill—the law filled the gap.

The plaintiffs sought to make the Class A Director a “constituency director,” a concept Delaware law has rejected in the corporate context.^[40] A director with contractual duties to act only in the best interest of the Class A stockholders would be placed “in the impossible position of serving two masters: The Class A Director would simultaneously owe a contractual obligation to pursue the best interests of the Class A stockholders plus a fiduciary obligation to pursue the best interests of the Company and all of its stockholders.”^[41]

Vice Chancellor Laster then applied the English law approach to implied terms to reject the plaintiffs’ efforts to show that the Governance Agreement contained a gap by failing to specify the duties of the Class A Director:

Borrowing from English law, the parties to the original negotiation would not regard a constituency-director provision as necessary to give business efficacy to their agreement, nor as so obvious that “it goes without saying.” If an officious bystander observing the negotiations proposed the provision, the parties would not respond with a loud, “Of course!” They would recognize that the provision could create problematic conflicts and necessitate a suite of additional contractual workarounds.^[42]

The Class A Stockholders Had No Implied Veto Right

The Governance Agreement contained an explicit suite of protective provisions, all based on the requirement that the Class A Director approve the listed actions. However, the Class A stockholders had not bargained for approval of those actions by the Class A stockholders. The Class A stockholders’ efforts to suggest that their failure to bargain for stockholder consent rights created a gap that needed to be filled were rejected. But for the sake of argument, Vice Chancellor Laster assumed there was a gap and then asked whether “it should be filled.”^[43] 

And again, Vice Chancellor Laster looked to the English law approach to implied terms for his answer:

[I]magine the original bargaining position and ponder whether, if the issue had come up at that time, the parties would have readily agreed to the specific vetoes the plaintiffs now want. It is not reasonably conceivable that the other parties would have readily agreed. Having granted the Class A protections, they would almost certainly want something in return for conferring additional rights on the Class A stockholders. To the same effect, if an officious bystander observing the negotiation suggested clarifying the [Governance] Agreement to require specific Class A stockholder approval, the parties would not respond with a united, “Of course!” Although the plaintiffs might have endorsed that view, the other parties would not. They would say something like, “You already have these protections, so why do you need more?” At best, further negotiation would have ensued. It is therefore not reasonably conceivable that the implied covenant can support the implicit veto rights.^[44]

The Implied Covenant’s Application to Discretionary Rights

Although “a party can wield a discretionary contractual right to protect its own interests[,]”^[45] the implied covenant limits the exercise of that discretionary right so that it cannot be used “maliciously and without contractual justification.”^[46] And again, Vice Chancellor Laster turned to the “officious bystander test” to illustrate why this must be so:

Given their agreed common purpose, a party in the original bargaining position would expect that the counterparty would not use a discretionary right to destroy the contractual relationship maliciously and without any justification rationally related to the shared contractual purpose. That premise is so basic that asking for a commitment against malicious action would be unthinkable. Framed from the English law perspective, a party to the negotiation who raised the issue would be met with the response that “it goes without saying.” If an officious bystander observing the negotiation suggested confirming that the contract prohibited a party from using a discretionary right to destroy the contractual relationship, both sides would immediately say, “Of course!”^[47]

But here, the plaintiffs could not show that the discretionary rights had, in fact, been exercised “maliciously and without any justification rationally related to the parties’ shared contractual purpose.”^[48] The company needed financing, and the defendants exercised their discretion in favor of the Class F Financing over other financing options. The fact that other financing options may have offered better opportunities for returns to the Class A stockholders did not appear to factor into the implied covenant analysis—that might await a subsequent decision on the fiduciary duty claims.^[49] But the implied covenant as applied to the exercise of discretionary rights only required the exercise of those rights to have “a contractually grounded reason for pursuing the Class F Financing.”^[50] Thus:

Facing a situation where the Company could fail and the investors end up with nothing, the defendants could properly exercise their contractual rights to obtain the financing the Company needed. It is not inferable that the defendants approved the Class F Financing for the sole purpose of harming the Class A stockholders, so the claim based on the discretionary-exercise version of the implied covenant fails.^[51]

Concluding Observations

I am currently working with coauthors on a proposed article with the working title “Making Sense of the Implied Covenant of Good Faith and Fair Dealing.” Based on informal surveys of academics and practitioners, there appears to be a general lack of understanding of the implied covenant, and I think the courts are confused at times as well. Vice Chancellor Laster’s recent opinion in Footprint International is a helpful step in providing some clarity on the extremely limited nature of the implied covenant, at least in Delaware.^[52]

Indeed, the implied covenant of good faith and fair dealing in Delaware may be a misnomer. There does not appear to be any overriding covenant implied in contracts in Delaware that requires contracting parties to act in good faith or with fair dealing. Instead, the implied covenant is simply a means of implying terms in a contract that are necessary to give full effect to the parties’ expressly agreed terms—i.e., “[t]he doctrine . . . operates only in that narrow band of cases where the contract as a whole speaks sufficiently to suggest an obligation and point to a result, but does not speak directly enough to provide an explicit answer.”^[53]

As one commentator has suggested, “[t]o display good faith in contract performance is simply to recognize the authority of the contract, and hence the authority of one’s counterparty to insist on performance according to the contract’s terms.”^[54] And those terms include both the express terms and those implied terms necessary to give effect to the express terms and which “are so obvious . . . that the drafter would not have needed to include the conditions as express terms in the agreement.”^[55] Thus framed, is the implied covenant of good faith and dealing all that different from the English law approach to implied terms generally?^[56]

And if you are curious about what may have prompted Vice Chancellor Laster to look to the English decisions on implied terms as a helpful guide to implying terms to fill gaps for purposes of the implied covenant of good faith and fair dealing in Delaware, he credits “nerdy discussions” with a “learned commentator on contract law” that were “unconnected to this or another case.”^[57] If you are curious who that might have been, see footnote 78 of Vice Chancellor Laster’s opinion in Footprint International.^[58]

1. [1939] 2 KB 206, 227. ↑

2. No. 2024-0968-JTL, 2026 WL 1180159 (Del. Ch. Apr. 30, 2026). ↑

3. Except with respect to contracts governed by the Uniform Commercial Code and those involving “special relationships,” the Texas Supreme Court has explicitly “rejected the argument that [it] should imply into contracts a covenant that would require the parties not to do anything that injures the right of another party to receive the benefits of the agreement.” Barrow-Shaver Resources Co. v. Carrizo Oil & Gas, Inc., 590 S.W.3d 471, 490–491 (Tex. 2019); see also English v. Fischer, 660 S.W.2d 521, 522 (Tex. 1983) (adopting such a “novel concept [of good faith and fair dealing] . . . would abolish our system of government according to the settled rules of law and let each case be decided upon what might seem ‘fair and in good faith,’ by each fact finder.”). But is the version of the implied covenant as adopted in Delaware the same version that Texas rejected? ↑

4. Restatement (Second) of Contracts § 205 (Am. L. Inst. 1981). ↑

5. Johnson & Johnson v. Fortis Advisors LLC, 352 A.3d 229, 255 (Del. 2026). ↑

6. Id. at 251. ↑

7. Id. at 254. ↑

8. Id. at 259. ↑

9. Id. at 253–54. ↑

10. Guilbeau v. Footprint Int’l Holdco, Inc., No. 2024-0968-JTL, 2026 WL 1180159, at *2 (Del. Ch. Apr. 30, 2026). ↑

11. Id. ↑

12. Id. ↑

13. Id. at *8. ↑

14. Id. ↑

15. Id. at *10. ↑

16. Id. ↑

17. Id. at *19. ↑

18. Id. at *10. This decision did not resolve all the claims made by the plaintiffs. See id. at *1 n.1. The court separately denied the defendant’s motion to dismiss the plaintiffs’ breach of fiduciary duty claims. See Guilbeau v. Footprint Int’l Holdco, Inc., No. 2024-0968-JTL, 2026 WL 1329169 (Del. Ch. May 11, 2026). ↑

19. Footprint Int’l, 2026 WL 1180159, at *10. ↑

20. Id. at *11. ↑

21. Id. at *10. ↑

22. Id. ↑

23. Id. at *11 (quoting New Enter. Assocs. 14, L.P. v. Rich, 292 A.3d 112, 138 (Del. Ch. 2023)). ↑

24. See Symbiont.IO, Inc. v. Ipreo Holdings, LLC, No. 2019-0407-JTL, 2021 WL 3575709, at *31 (Del. Ch. Aug. 13, 2021), discussed in Glenn D. West, When Is a Person’s Status as an Affiliate Relevant?, Weil Glob. Priv. Equity Watch (Sept. 9, 2021). ↑

25. See Symbiont.IO, 2021 WL 3575709, at *31. ↑

26. Id. ↑

27. Another example is where a charter or limited liability company (“LLC”) agreement prohibits amendments that would adversely impact the rights and preferences of a stockholder or LLC member, but does not specifically prohibit those consequences from being obtained through a merger. The doctrine of “independent legal significance” may well permit a merger to strip those rights even when a direct amendment would require the consent of the affected stockholders or members. See Elliott Assocs., L.P. v. Avatex Corp., 715 A.2d 843, 855 (Del. 1998) (“The path for future drafters to follow in articulating class vote provisions is clear. When a certificate . . . grants only the right to vote on an amendment, alteration or repeal, the preferred have no class vote in a merger. When a certificate . . . adds the terms ‘whether by merger, consolidation or otherwise’ and a merger results in an amendment, alteration or repeal that causes an adverse effect on the preferred, there would be a class vote.”). Even though it has been suggested otherwise, the implied covenant does not appear to treat the failure to specify a merger in such circumstances as a gap to be filled. See D. Gordon Smith, Independent Legal Significance, Good Faith, and the Interpretation of Venture Capital Contracts, 40 Willamette L. Rev. 825 (2004); David I. Albin & Cole Mayhew, USA-Connecticut Trends and Developments: Recent Developments on the Implied Covenant of Good Faith and Fair Dealing and Their Implications for Connecticut Law, Chambers & Partners Prac. Guide: Corp. M&A 2026 (Apr. 21, 2026). ↑

28. Footprint Int’l, 2026 WL 1180159, at *13. ↑

29. Id. ↑

30. McKenzie v. BDO USA, P.C., No. 2025-0264-LWW, 2026 WL 191010, at *5 (Del. Ch. Jan. 26, 2026) (citations omitted). But depending on the facts, merely using the term “sole discretion” may not be sufficient to avoid the gap-filling imposition of the implied covenant to constrain the exercise of that discretion. See Glenn D. West, Musings on the Exercise of “Sole Discretion,” Weil Glob. Priv. Equity Watch (Aug. 29, 2022). ↑

31. Footprint Int’l, 2026 WL 1180159, at *22. ↑

32. See Yam Seng Pte Ltd. v. Int’l Trade Corp. Ltd., [2013] EWHC 111 (QB), at paras. [119]–[154]; Bristol GroundSchool Ltd. v. Intelligent Data Capture Ltd., [2014] EWHC 2145, at paras. [175], [196]; Ellis v. John Benson Ltd., [2025] EWHC 2096 (KB), at paras. [249]–[265]; see also Maud Piers, Good Faith in English Law—Could a Rule Become a Principle?, 26 Tul. Eur. & Civ. L. F. 123 (2011) (discussing the then current state of the law in England regarding the principle of good faith); David B. Kierans & Julia Kappler, Good Faith in Canadian Contract Law, Bus. L. Today (Apr. 20, 2016) (discussing both Canadian and UK law). For a more recent take on the Canadian perspective, see Brandon Kain, Marina Sampson & Foti Vito, Recent Developments in the Law of Good Faith—Key Takeaways, McCarthy Tetrault Insights (May 4, 2026). ↑

33. Footprint Int’l, 2026 WL 1180159, at *14. ↑

34. Id. (quoting BP Refinery (Westernport) Pty Ltd v. Shire of Hastings (1977) 180 CLR 266, 282–83). ↑

35. Id. (quoting Shirlaw v. S. Foundries (1926) Ltd., [1939] 2 KB 206, 227). ↑

36. Id. (quoting Marks & Spencer plc v. BNP Paribas Sec. Servs. Tr. Co. (Jersey) Ltd., [2015] UKSC 72, at paras. [21], [23]). ↑

37. Id. at *17, *19, *22. ↑

38. Dieckman v. Regency GP LP, 155 A.3d 358, 361 (Del. 2017). ↑

39. Footprint Int’l, 2026 WL 1180159, at *15. ↑

40. Id. In most private equity deals, the holding company would have been an LLC, not a corporation, and while the implied covenant would still apply, the parties could most likely have members exercising the approval rights with all fiduciary duties waived. See Del. Code § 18-1101(c) (“To the extent that, at law or in equity, a member or manager or other person has duties (including fiduciary duties) to a limited liability company or to another member or manager or to another person that is a party to or is otherwise bound by a limited liability company agreement, the member’s or manager’s or other person’s duties may be expanded or restricted or eliminated by provisions in the limited liability company agreement; provided, that the limited liability company agreement may not eliminate the implied contractual covenant of good faith and fair dealing.”). ↑

41. Footprint Int’l, 2026 WL 1180159, at *16. ↑

42. Id. at *17. ↑

43. Id. at *19. ↑

44. Id. ↑

45. Id. at *22. ↑

46. Id. ↑

47. Id. ↑

48. Id. ↑

49. Id. at *1 (“The court will issue a separate decision addressing the fiduciary and related claims.”). And on May 11, 2026, the court did issue a separate decision regarding the fiduciary duty claims. See Guilbeau v. Footprint Intern’l Holdco, Inc., C.A. No. 2024-0968-JTL, 2026 WL 1329169 (Del. Ch. May 11, 2026). In denying the motion, in part, to dismiss the plaintiffs’ claim for breach of fiduciary duty against five of the ten directors of the company, Vice Chancellor Laster held that:

    Five of the ten directors inferably faced a conflict of interest for purposes of the Class F Financing. The plaintiffs therefore succeeded in rebutting the business judgment rule. Entire fairness inferably applie[d] to the Class F Financing.
    . . .
    The Class F Financing was inferably unfair. It was inferably coercive and did not offer all stockholders an opportunity to participate on the same terms. The plaintiffs have stated a claim for breach of fiduciary duty under the entire fairness test.

    Id. at *26 and *32. ↑

50. Footprint Int’l, 2026 WL 1180159, at *22. ↑

51. Id. ↑

52. Footprint International follows another recent decision by Vice Chancellor Laster addressing the implied covenant in Delaware, Calumet Capital Partners LLC v. Victory Park Capital Advisors, LLC, 353 A.3d 88 (Del. Ch. 2026). ↑

53. Airborne Health, Inc. v. Squid Soap LP, 984 A.2d 126, 146 (Del. Ch. 2009). ↑

54. Daniel Markovits, Good Faith As Contract’s Core Value, 2021 Mich. St. L. Rev. 1, 18 (2021). ↑

55. Dieckman v. Regency GP LP, 155 A.3d 358, 361 (Del. 2017). ↑

56. The English approach to implying terms has even been described by one commentator as providing “gap fillers.” See Piers, supra note 32, at 157. One may well additionally ask whether, if the implied covenant of good faith and fair dealing (under Section 205 of the Restatement (Second) of Contracts) has the limited meaning ascribed to it by the Delaware courts, is it really that different from the concept of supplying an omitted term necessary to give effect to the contract under Section 204 of the Restatement (Second) of Contracts. See Restatement (Second) of Contracts § 204 (Am. L. Inst. 1981) (“When the parties to a bargain sufficiently defined to be a contract have not agreed with respect to a term which is essential to a determination of their rights and duties, a term which is reasonable in the circumstances is supplied by the court.”). Moreover, did Texas reject a different version of the implied covenant of good faith and fair dealing than the version currently being applied in Delaware? See supra note 3. ↑

57. Footprint Int’l, 2026 WL 1180159, at *14 n.78. ↑

58. Id. ↑

At the recent Tokyo Business and Rule of Law Forum, Alejandro Ponce, Executive Director of the World Justice Project (“WJP”), issued a stark warning: “A weak rule of law is a direct business problem. When everyone—including government—is not held accountable to clear, equally applied rules, the private sector loses the confidence required for long-term planning.”

The data confirms instability is accelerating. According to the WJP Rule of Law Index^® 2025, 68% of countries saw their scores decline this year—a significant jump from 57% in 2024. While building stable institutions is a slow, iterative process, the current trend shows that dismantling them occurs much more swiftly.

As lawyers, who ourselves have taken a solemn oath to defend the Constitution and the rule of law, we must heed the warning that the erosion of judicial independence and civil justice represents a critical threat to the democratic and economic stability we are sworn to protect.

A rise in authoritarianism is driving the rule of law recession

An expansion of authoritarian trends—namely, a reduction in civic space and weakening checks and balances—has been the primary force behind this downturn, with deep declines in factors measuring Constraints on Government Powers, Open Government, and Fundamental Rights.

The critical pillars of government accountability have eroded in most countries:

• Legislative and judicial oversight are losing their ability to check executive power, declining in 61% of countries.
• Independent auditing of government actions fell in 61% of countries.
• Fundamental freedoms, specifically expression, assembly, and participation, have shrunk in nearly 75% of nations, stifling the public debate necessary for a healthy democracy.

Judicial independence, the last line of defense against executive overreach, is weakening

The Index shows that judiciaries are losing ground to executive overreach, with rising political interference across justice systems. Indicators measuring whether the judiciary limits executive power and whether civil and criminal justice are free from improper government influence declined in 61%, 67%, and 62% of countries, respectively.

For the business community, the decline of civil justice in 68% of countries is particularly concerning. This weakening is reflected in longer court delays, less effective mediation, and increased government interference in legal outcomes.

Rule of law in the U.S. continues to decline

Rule of law in the United States continued to decline for the third consecutive year, falling in seven of the eight factors measured. Indeed, in the latest report, the United States was one of the ten countries with the largest declines in rule of law. This decline was driven specifically by weakening fundamental rights and a lack of effective constraints on government powers, as well as the perception that government officials are not sanctioned for misconduct. The U.S. currently ranks twenty-seventh out of the 143 countries and jurisdictions surveyed.

In the latest World Justice Project report, the United States was one of the 10 countries with the largest decline in rule of law.

U.S. rule of law saw the sixth largest decline of the 143 countries assessed, behind only the Russian Federation, Sudan, Mozambique, Togo, and Mexico. Source: © World Justice Project, WJP Rule of Law Index 2025.

Seven of eight rule of law factors declined in the United States from 2024 to 2025.

U.S. scores for seven of the eight rule of law factors dropped by between 0.01 and 0.04; only Order and Security showed no change. Source: © World Justice Project, WJP Rule of Law Index 2025.

The continuing decline in the U.S. rule of law score was driven specifically by weakening fundamental rights and a lack of effective constraints on government powers.

From 2016 to 2015, U.S. scores for fundamental rights and constraints on government powers dropped from 0.75 to 0.65 and 0.81 to 0.63, respectively. Source: © World Justice Project, WJP Rule of Law Index 2025.

The U.S. score for the subfactor “Government officials are sanctioned for misconduct” continued to decline.

The United States scored 0.52 for the subfactor Government Officials Are Sanctioned for Misconduct, down from 0.69 in 2016. Source: © World Justice Project, WJP Rule of Law Index 2025.

WJP Rule of Law Index 2025: Key Findings Summary

The WJP Rule of Law Index provides data on eight factors: Constraints on Government Powers, Absence of Corruption, Open Government, Fundamental Rights, Order and Security, Regulatory Enforcement, Civil Justice, and Criminal Justice. Scores range from 0 to 1, where 1 signifies the highest possible adherence to the rule of law. This year, the Index covers 143 countries and jurisdictions.

• Top-Ranked Countries: Denmark (1), Norway (2), Finland (3), Sweden (4), New Zealand (5).
• Bottom-Ranked Countries: Venezuela (143), Afghanistan (142), Cambodia (141), Haiti (140), Nicaragua (139).
• Top Decliners: The most significant decliners include the Russian Federation (-4.9%), Sudan (-4.4%), and Mozambique (-3.9%).
• Top Improvers: Countries with the largest improvements include the Dominican Republic (2.1%), Senegal (1.6%), and Sierra Leone (1.4%).

Explore the full rankings and findings of the 2025 WJP Rule of Law Index at the World Justice Project website.

“These findings and data are important for both businesses and business lawyers,” said John H. Stout, of counsel at Fredrikson & Byron P.A. and co-chair of the American Bar Association Business Law Section’s Rule of Law Working Group. “This information is critical to both as they give legal advice and make business decisions about whether to increase or decrease their investment in a particular country.”

About the World Justice Project:

The World Justice Project (WJP) is an independent, nonpartisan, multidisciplinary organization working to create knowledge, build awareness, and stimulate action to advance the rule of law worldwide.

This article is part of a series on the rule of law and its importance for business lawyers created by the American Bar Association Business Law Section’s Rule of Law Working Group. Read more articles in the series.

Some people just seem to know instinctively where the career pathways can be found in the legal profession, and how to travel them successfully. But I did not, and it’s possible that you may not either. If that sounds familiar, and if you have ever wondered about whether a clerkship for a federal judge might be for you, then please keep reading so that I can share some personal reflections and practical tips on the who, what, when, where, why, and how of becoming and serving as a law clerk to a federal judge.

First, let me provide some context. I am a federal judge and sit in the United States Bankruptcy Court for the Eastern District of New York, in Brooklyn, NY. Our court is a unit of the federal district court, and like all bankruptcy courts, we handle all of the bankruptcy cases that are filed in the district, from the biggest corporate Chapter 11 case to the individual case of the person who is just trying to get a fresh start in a Chapter 7 or 13 case. All bankruptcy cases are federal, and all bankruptcy issues come to our court.

I generally have three law clerks, with some serving in two-year term law clerk positions and others serving in one-year temporary law clerk positions. In the past, I have also had a career law clerk from time to time; this is a phrase we use in the federal system for someone who stays with a judge from year to year. I also hire two to three term law clerks each year. I typically hire at least one of those law clerks in the fall, to begin one year later, in the following fall. I may hire one or two more law clerks depending on the caseload and the level of available funding. Sometimes those additional positions are partially paid, sometimes those positions are fully paid, and sometimes those law clerks work as volunteers with stipends from their law schools, or, if they are at a firm, with the expectation of receiving a public service bonus when they finish.

So, with that background, let’s dive into the who, what, when, where, why, and how of serving as a law clerk to a federal judge.

Who should consider applying to be a law clerk to a federal judge?

Let’s start with who might consider seeking a federal judicial clerkship. Too often, it may seem that these positions are for someone else—someone with better grades, law review experience, connections with faculty, things like that. But there are as many paths to these positions as there are federal judges, and even within a particular judge’s chambers, most judges consider a broad range of criteria in assessing clerkship applicants.

So, if you are drawn to the idea of working full-time in the federal judiciary, in a judge’s chambers, with the opportunity to do the work of the justice system on a daily basis, then you—yes, you—should consider applying to be a law clerk to a federal judge.

What does a law clerk to a federal judge do?

Every chambers is different, but still, there are some common themes in the work of a law clerk to a federal judge. My law clerks help me with absolutely all of the work of chambers, including researching motions, drafting bench memos, attending court hearings, and drafting orders and decisions. For me, court hearings and the work that comes before and after them, including bench memos and preparing orders, are a big part of what we do. I could have thirty or more matters on a given day’s calendar, and I will decide most of these from the bench. A law clerk helps me to prepare for each and every one of these matters and assists with drafting the order after the matter is decided, based on our review of the record, our knowledge of the case—including the bigger picture, and the arguments we hear that day.

In the bankruptcy world, we spend a good deal of time in the courtroom, and we tend to see our cases in the courtroom much more frequently than in civil litigation in the district court. Whether the matter is a very complex summary judgment motion in a $500 million adversary proceeding or a motion to dismiss brought by a debtor in their own Chapter 7 bankruptcy case, a bench memo of some sort is prepared.

One of my law clerks always accompanies me in the courtroom while I am on the bench—this can be very helpful when an unexpected question comes up. And sometimes during a hearing, it is helpful to have the law clerk’s perspective and judgment on how to decide a matter or manage a situation. This can be some of the most interesting work done by the law clerk, and it can be a very special learning experience as well. You learn so much from spending dozens of hours in a courtroom, seeing how the legal system can work when it works at its best, and appreciating the role of the lawyers and the court in making that happen. As the law clerk observing proceedings, you appreciate how lawyers can do their work very well indeed, and you also see situations where you will realize that you could do a better job. No matter what, it is our job to treat people with the utmost respect and make them feel, win or lose, that they have never been treated better. And of course, every case is our most important case.

We also try to think about how we can make the system work as well as possible. In this respect, my law clerks help me with work we do in connection with pro bono organizations that provide lawyers and support for people who can’t afford counsel. I teach a seminar at one of the New York area law schools, and sometimes I look to my law clerks for teaching ideas since they are a lot closer to the law school experience than I am. My clerks also help me prepare for my work with judges from other countries; from time to time they have assisted me with hosting delegations from Europe, South America, the Middle East, and China. They work with our interns, who come from U.S. law schools and universities, and also from law schools outside the United States. The clerks learn how to supervise and mentor our interns, and they discover that they are ready to be mentors, even just one year out of law school. And finally, my law clerks assist with every other thing that needs to be done in chambers, whatever it may be. They help me get through a long day, brainstorming with everyone in my chambers on how to get our work done. There is simply nothing that I do as a judge that my law clerks don’t help me with.

I often hold settlement conferences in my own cases, and occasionally, I serve as a mediator in cases that are referred to me by my judicial colleagues. Sometimes my law clerks will assist with that work as well. I like the clerks to have that experience because this type of problem-solving in dispute resolution is a significant part of what lawyers do, and more and more, a significant part of what judges should do. Indeed, every spring, my law clerks attend a week-long commercial mediation training program offered by the New York State Bar Association. When the clerks return from that training, they have gained new skills in seeing a dispute from the perspective of each of the parties, identifying both positions and interests, and managing a case in a way that permits, and even promotes, the parties reaching a resolution. So on a typical day, they might spend some time in court, some time in chambers, some time with the interns, and some time with me.

What does this look like from the law clerk’s perspective? I hope that for my law clerks, a typical day is a day where they know that they are going to make a difference. They will likely spend some time in the courtroom, prepare files, talk through issues, interact with interns, help solve an unexpected case issue, and coordinate with me and our extraordinary courtroom deputy, who serves as the chambers interface with the bar. We all prepare as best we can, and we all must be prepared for the unexpected as well.

When does it make sense to pursue an application to be a law clerk to a federal judge?

Many law students think about pursuing a clerkship directly after law school, and that can make a lot of sense. After all, clerking can be a very logical and natural transition from legal education to law practice. Law schools provide support for clerkship applicants, and professors who may serve as mentors, recommenders, and coaches may be just down the hall.

Others may plan to apply for clerkships after a few years of private practice—and indeed, many judges specifically seek out applicants with that kind of practical professional experience. To be sure, it can be challenging to interrupt your practice path for what may seem like a detour. But your experience in practice may well make you a stronger applicant, and a more effective law clerk, too.

The fact is, there is no wrong time to pursue an application to be a law clerk to a federal judge. And if you apply for the year directly following law school and are not successful, you can apply again—and again after that, if it makes sense to do so. The experience that you gain as a law clerk, and even the experience that you gain as an applicant, will help you sharpen your thinking about the law and the justice system and your goals and aspirations in the legal profession. To paraphrase a well-known proverb, while it may seem like the best time to apply for a federal judicial clerkship was last year, or last month, or yesterday, the second-best time is . . . now.

Where should I consider applying for clerkships?

One of the most engaging and least appreciated aspects of serving as a law clerk to a federal judge is that you can do it just about anywhere—from a major metropolis to some national parks. Judges generally do not expect you to be admitted to the bar of the state in which they sit. It can be very helpful if you are familiar with the local legal community—but again, judges do not expect that you will be. You may well choose to seek a clerkship in the city in which you plan to practice. At the same time, a clerkship gives you an opportunity to immerse yourself in a part of the country that is brand-new to you. It may make sense to apply for clerkships in the locale where your law school is located, or where many of its alumni practice. But many other candidates will likely be doing the same thing, so if you seek to distinguish yourself, you should consider whom you would like to work for, and where you might want to live. It’s helpful to be prepared to explain why a particular court and locale appeal to you, including in your cover letter. This is especially so if you do not otherwise have a particular connection to the city or region.

So, think big! Think broad! Be willing to think outside the box! If you are excited about the opportunity to work in that judge’s chambers and live in that community, then you should not hesitate to apply for that position.

Why should I consider applying to be a law clerk to a federal judge?

When I started my clerkship with a district judge, he looked down at me, a full foot and half shorter than he was, and said, “You now have the second-best job a lawyer can have. You are a law clerk to a federal judge.” He then patted himself on the chest and said, “I have the best job. I am a federal judge.” In a nutshell, if you like it, you will love it; it truly is one of the best jobs you can have in the law. You will learn and discover things from your clerkship that will assist you for the rest of your career. I still discover things I learned as a law clerk in the mid-1980s that inform me, inspire me, and make me want to skip on the way to work and sing on the way home. Being a part of the federal courts is truly a privilege, the best kind of public service, and a great job, too.

On a practical level, as a law clerk, you will learn how to synthesize information quickly and get to the point—and that is an invaluable skill for a practicing lawyer. At the same time, you will learn not to leap to conclusions but to assess a file and form a judgment quickly, carefully, and confidently so that you are able to defend it when questioned. If you work for a judge who is in the courtroom a lot, you will see fabulous lawyering and you will learn from watching that. You will also see, hopefully not too often, lawyers who will make you think, “I can do this better than they can do it—right now.” Knowing how to be comfortable when you walk into a courthouse or courtroom is a wonderful skill, and the earlier in your career that you can learn it, the better. You will read great briefs and not so great ones. And you will learn through experience that many people do not always give a final read-through to the documents that they file—and hopefully, you will never make that mistake yourself!

And you will form a lifelong relationship with your judge, who will be your mentor, supporter, cheerleader, reference, and sounding board for life. We like to say in our chambers that “chambers is forever,” and as a law clerk, you become a part of a chambers family. You will have the benefit of that chambers experience and those relationships for your entire professional life. As just one example, I recently had the opportunity to administer the oath of office as a New York State Court of Claims judge to one of my former law clerks, who is now serving as a judge. I can’t think of a more special moment to share with a former law clerk than to see her become a judge herself, and I could not have been more proud of her.

Of course, you will also learn a good deal of substantive law, in more areas than you can imagine. In this court in particular, you will get smart about deals, and you will get smart about litigation. You will identify problem-solving as a big part of being a lawyer, which will forever make your clients grateful. And you will have a lot of fun.

Another reason to consider a clerkship is that it may well open doors for you, in the near term and down the road, too. Again, looking to my own law clerks as examples, many have gone on to practice in law firms, from the largest global “big law” firms to boutique specialty firms, and everything in between. Several have gone on to do other clerkships with other federal courts, and in the state court as well, including the Commercial Division of the New York Supreme Court in Manhattan, one of the preeminent business courts in the country. Several of my former law clerks have gone into public interest work and public service. One was the first bankruptcy court law clerk to be selected to be a Supreme Court Fellow. At least one is a journalist. And as noted, one former law clerk is now a judge, and I suspect that others will follow.

How do I apply to be a law clerk to a federal judge?

Our hiring process begins with the “OSCAR System,” the web-based resource that the federal courts use for the electronic submission of clerkship applications to federal judges. Many—perhaps most—federal judges use this system to receive applications. If you have identified a particular judge to whom you would like to apply and do not see the judge listed on OSCAR, it may make sense to contact chambers to determine whether the judge is currently considering applications and how they would like to receive them.

In my hiring process, we rely on OSCAR and also occasionally receive applications by mail, both electronic and paper. We encourage everyone to apply via OSCAR, and ask for applicants to submit a cover letter, a résumé, college and law school transcripts, a writing sample, and references. We receive many applications for each position, review them carefully, and call a small number of individuals in for interviews.

An interview is generally a very important part of the clerkship hiring process. You should expect to interview with the judge and also with the judge’s current team of law clerks, and each of these interviews matters. In my chambers, when you interview to be my law clerk, it’s a half-day, in-person enterprise. And everyone in chambers is involved. You will interview with me of course, and also with each of the current law clerks. References can be very important, as the judge will rely on them to get a better sense of the applicant from people who have taught or worked with them for a period of time. Deciding who to hire as my law clerks is probably the single most important decision I make every year because they are my chambers team and we depend on each other enormously.

What do federal judges look for in their law clerks? There are some common denominators. Academic performance and consistency are important, of course. But law school grades are not the only thing that matters. Like many judges, I look for trends in an applicant’s academic performance, as well as how well the applicant has done. If you are getting B’s and C’s, rather than A’s and B’s, it is going to be difficult to be considered seriously for a law clerk position. But you may have a great opportunity to serve as an intern for a federal judge, and that may then lead to other opportunities.

Course selection is also something to consider, and some law school courses may prove to be especially helpful to an aspiring law clerk. One is civil procedure, as well as any advanced procedure classes that may be offered. Federal courts and federal jurisdiction are also important subjects and may help to demonstrate your interest in courts and court processes. And in light of the problem-solving nature of trial courts, including district and bankruptcy courts, a negotiation or mediation course may provide helpful perspective as well. Most important, especially in the second and third year, is to take the classes you love, that you are excited about, that challenge you to do your best and most critical thinking and writing, and that remind you of all the reasons you wanted to be a lawyer. In the end, any class that helps you learn to write, research, think, and exercise judgment as a functioning, articulate professional is helpful.

Speaking for myself, I want to meet and consider the people who have done more than just do very well in undergraduate and law school, and the people who have done things in addition to just going to school, whether before or after their law school education. In terms of experience, I have hired law clerks directly from law school, I have hired law clerks who are completing clerkships with other judges, and I have hired law clerks who have practiced for several years.

As I consider applications, I try to get a sense of the person’s life story and why they are interested in serving as a law clerk. From their cover letters, résumés, recommendations, courses, and law school activities, I try to get a sense of whether this is someone who will love the work of our court, being a part of our chambers, and feeling that public service is truly important. I look for applicants who like the idea of going to work every day and making a difference in the justice system, in the work of companies large and small, and in the lives of ordinary people who are trying to get a second chance at life. If I get a sense that the law clerk prospect is someone who will love that, and who is interested in the position for all of those reasons, that is probably someone I will want to meet and interview.

Finally, don’t rule out applications to courts outside the federal system. In many of the state courts, including in New York, there are court attorneys who work from year to year supporting the judges in their chambers. These are very interesting career positions and sometimes are a pathway to the state court bench. Similarly, in some federal courts of appeal, there are staff attorney law clerks who work with the entire court.

There are also other roles within the court system for attorneys. For example, we have a pro se attorney who works several hours a day with an “open door” office for pro se access; our court’s pro se attorney was one of the first in the country and has become a model for other courts. Court administration, including the clerk of court, chief deputy, and district and circuit executive’s offices, can also provide a very interesting public service career path.

* * *

Finally, here are some parting thoughts for anyone who is considering applying for a federal judicial clerkship, or any other dream position in the legal profession—because serving as a law clerk, or as a judge someday down the road, is truly a dream job.

First, do not let anyone tell you that you cannot do it. Put another way, don’t reject yourself! Be sure to do as well as you can in law school, because your grades matter. Remember to tell your story, to share what excites you, and remember too why you went to law school. Studying is important, of course—but is not the only important thing. Try to stay involved with all of the things that you love to do in your community.

Next, don’t worry if you aren’t perfect—no one is. I have had wonderful law clerks who were on law review, and I have had wonderful clerks who were not on law review. I have had wonderful law clerks who had perfect GPAs from their first semester of law school, and I have had wonderful law clerks whose transcripts told me that the first semester was hard for them, but they figured it out, and excelled in the end.

In addition, be open to relocating to a different part of the country. You will parachute into that legal community in a very engaging way, and you will be part of the federal court. You will immediately get to know the bar. You will be a part of the chambers community, which is like a little family. And who knows, maybe you will live there for the rest of your professional career!

Be open to possibilities. I was among the least likely bankruptcy court judge applicants you can imagine—I had no experience in bankruptcy court and had never practiced bankruptcy law—but something about this “second-chance” court seemed very special to me, and I was lucky enough to be appointed.

And finally, always have faith in yourself. Here’s one more story to illustrate that. I still remember one late evening many years ago—decades ago, actually!—I was applying for scholarships to help pay for college. One seemed especially unlikely, and I was on the verge of abandoning the application. But my mother quietly said, “They have to pick someone, and it might as well be you!” Figuring that I had nothing to lose, I applied for that unlikely scholarship —and I got it.

So, if you think that a federal judicial clerkship is for you, apply! The judge has to pick someone, and it might as well be you!

This article is related to a CLE program that took place during the ABA Business Law Section’s 2026 Spring Meeting. To learn more about this topic, listen to a recording of the program, free for members.

Effective April 18, 2026, the Nationwide Multistate Licensing System (“NMLS”) implemented changes to the Individual (MU2 and MU4) Disclosure Questions along with definitional updates that also apply to Company (MU1) and Branch (MU3) Disclosure Questions that will impact existing licensees (“Update”). The Update requires disclosure question updates for individuals identified as control persons, branch managers, licensed Mortgage Loan Originators, and qualifying individuals who have MU2 profiles associated with a company’s license. The NMLS Glossary of terms was also updated to amend existing definitions and incorporate new definitions.

The NMLS is the system of record for non-depository, financial services licensing or registration for participating state or territorial governmental agencies, including the District of Columbia and U.S. Territories of Puerto Rico, the U.S. Virgin Islands, and Guam. In these jurisdictions, NMLS is the official system through which companies and individuals apply for, amend, renew and surrender license authorities managed through NMLS. These agencies require companies seeking licensure maintained through the NMLS, the individuals designated as control persons, branch managers, licensed Mortgage Loan Originators, and qualified individuals of such companies to answer Disclosure Questions pertaining to various topics, including criminal, civil, and regulatory items that may impact licensure, and make updates to maintain the accuracy of these answers on the NMLS.

The Update requires individuals with associated MU2/MU4 profiles to review their existing information compared to the new Disclosure Questions and the updated terms in the Glossary. For example, the existing term “financial services-related” is no longer applicable in the Individual (MU2/MU4) context, and a new term, “financial services,” was added to the MU2/MU4. However, the term “financial services-related” still applies in the Company (MU1) Disclosure Question context. Terms such as “found,” “order,” and “governmental entity” are equally applicable between the Company (MU1) and Individual (MU2/MU4) Disclosure Questions.

While the Update largely reformats and clarifies existing questions, this is not entirely the case, and the changes may in some instances implicate new disclosures. It is important to note in this regard that the revised questions are retroactive and that the scoping of some questions has been narrowed. If revisions are required due to the changes, the commentary describing the changes recommends not adding a new explanation, but rather updating the existing fields in the Explanation section of the Disclosure Questions where applicable. If there is any change in an existing response, the explanation should be updated accordingly. It is our understanding that any previously uploaded documents and associated explanations will be available to be “matched” to “yes” answers for the new disclosure questions.

The NMLS materials regarding the Update indicate an August 31, 2026, deadline for compliance, which is likely realistic for individual Mortgage Loan Originators or other individuals associated with companies that are not often making updates in the NMLS. However, the updates may be more time sensitive from the Company licensing perspective if a company has upcoming filings, given that the licensed entity will not be able to make any filings of any kind in the system until the Disclosure Questions for all individuals associated with the Company-level record are updated. The procedural consequence of the update requirements is that the updates essentially halt the licensee’s ability to respond to regulator inquiries via the NMLS or submit new license filings until all associated individuals have updated and attested to their individual filings.

The Conference of State Bank Supervisors has published materials on NMLS to assist with this process.