Can Soaring Tech Salaries Still Qualify for R&D Tax Credits?

A fundamental shift is underway. AI and big tech are engaging in a talent war, driving compensation packages for elite engineers into the seven or even nine figures. These aren’t just salaries; they are complex instruments of equity, milestone bonuses, and retention grants designed to secure scarce, high-impact talent. For CFOs and tax leaders, this new reality creates a critical challenge: How do you substantiate R&D tax credits under IRC §41 when the “wages” are driven by an individual’s market value and future potential, not just the hours they work?

Recent audit experience across the technology sector confirms this is no longer a theoretical risk; IRS examiners are actively targeting these high-wage claims with a heightened level of scrutiny. As a result, the traditional approach to documenting research credits is no longer sufficient. To navigate IRS scrutiny and manage financial reserves effectively, companies may need to enhance their audit readiness. The substantiating documentation and data would need to separately establish the immense value of the Highly Paid Individual (HPI) from the qualified research activities they perform.

The Market Has Changed: Compensation Is Now an Asset Class

The numbers are staggering. Reports, such as in Forbes and other publications, indicate that average compensation for AI professionals has soared into the millions in many cases, driven by equity packages that dwarf historical tech benchmarks. This hyper-competition reflects a new understanding of talent: Elite engineers are not just labor, but sovereign-like assets who possess immense bargaining power and are central to value creation.

Critically, this compensation is delivered through a complex architecture of financial instruments, including, but not limited to:

• Multi-year equity grants (Restricted Stock Units) with extended vesting schedules.
• Performance-based equity contingent on achieving specific technical or product milestones.
• Substantial signing and retention bonuses designed to secure a multi-year commitment.

This compensation is designed to capture:

• Scarcity Value: There is a very limited pool of individuals who are seen to be most capable of building frontier AI models, and other advanced technology systems —a trend analyzed in global reports from organizations like the OECD and the World Economic Forum on the AI skills gap. Recent labor market analysis shows that demand for generative AI roles has surged by over 1,800%, underscoring the intensity of competition for top technical talent.
• Retention Risk: The cost of preventing a key individual from moving to a competitor.
• Reskilling Imperative: Executives estimate that 40% of their workforce will need to reskill in the next three years as a result of implementing AI.
• Future Option Value: This is a bet that one person’s insight could unlock billions in enterprise value. The compensation often functions like a venture capital investment in a person, a dynamic increasingly reflected in AI executive compensation studies. It is a payment for the possibility—the “option"—that one individual’s unique insight could lead to a breakthrough product, a fundamental efficiency gain, or a pivotal discovery that yields returns far exceeding the initial outlay.

This framing is essential, because when an IRS examiner sees a W-2 with millions in stock-based compensation, their first question won’t be about the market—it will be about the work.

The Tax Law Collision: High Pay vs. Qualified Services

The good news is that IRC §41 does not impose a cap on compensation. A multi-million dollar wage is just as eligible for the research credit as any other. Eligibility is, and has always been, determined by the nature of the services performed.

However, high compensation dramatically raises the audit stakes and the burden of proof. The IRS’s focus on research credit claims means examiners are trained to scrutinize:

• Activity vs. Title: Was the “Chief AI Scientist” performing hands-on experimentation, or were they focused on strategy, recruiting, and external evangelism?
• Contemporaneous Proof: Can you connect equity vesting or bonus payouts to specific, qualified research activities performed during that period?
• Allocation Methodology: How do you defensibly allocate compensation from complex, multi-year grants to mixed-duty roles that blend direct research with non-qualifying strategic or managerial tasks?

A weak narrative or thin documentation, when attached to a massive wage claim, presents a material risk of disallowance.

Enhanced Substantiation of Credit Claims

To defend these claims, companies must build a clear, logical bridge from the business value an individual represents to the specific qualified research they conduct. Acknowledging the “value” of an individual is not for justifying the credit itself, but for providing essential context for why the compensation is so high. Three layers of substantiation may help trace the “why”, “what” and “how” of the claim:

This model may help tell a more complete story along the lines of: “The potential to drive business value explains the high compensation. To do so, the individual delivered product output, which required them to conduct the following qualified research activities.”

Prepared for the Individual Audit? Key Questions for HPIs

When a multi-million-dollar wage claim is on the line, the audit lens may intensify on the individual. The questions move beyond “what did the company do?” to “what, specifically, did this person do every day?” The traditional four-part test for a business component, as outlined in the IRS’s Research Credit Claims Audit Techniques Guide (RCCATG), becomes mere table stakes. The real challenge is proving the HPI’s contribution with granular evidence.

As you consider your position, reflect on whether you can confidently answer the types of questions a tax professional would help you prepare for:

• Can you demonstrate how your HPI’s time and intellectual effort are allocated across the multiple projects and business components they influence? Can you defensibly map their complex compensation—from RSUs to bonuses—to specific qualified activities in a given year?
• Can you summarize a “day in the life” of your lead AI researcher? Could you provide a compelling narrative, supported by anecdotal evidence, that translates their abstract work into a concrete story of experimentation and qualified research for a non-technical audience?
• Would you be able to produce a comprehensive dossier of “engineering-grade artifacts” to corroborate that narrative on demand—e.g., patent applications, technical architectures considered, iterations of Proofs-of-Concept, activity logs or audit trails from generative AI-assisted coding platform, version control commits, and meeting summaries that meet the substantiation requirements of Treas. Reg. §1.41-4(d)?

These questions highlight a critical shift in the burden of proof. For HPIs, satisfying the R&D credit requirements for the project is necessary, but no longer sufficient. The audit may be won or lost on the ability to provide a verifiable, and compelling record of the individual’s daily qualified research activities. The focus is the person, not just the project.

The Strategic Takeaway

The core principle for navigating this new landscape is clear: The high value of an individual’s compensation package drives the exposure of an IRS audit, but only hard evidence of their research activities can prove eligibility for the tax credit. The era of the million-dollar engineer does not weaken the R&D tax credit; it demands a stronger, more disciplined approach to substantiation with engineering rigor.

This article does not necessarily reflect the opinion of Bloomberg Industry Group, Inc., the publisher of Bloomberg Law, Bloomberg Tax, and Bloomberg Government, or its owners.

Author Information

Ajay Wanchoo is a senior managing director in the Accounting Methods and Credit Services practice in NYC. Hogan Humphries and Alexander Fox are managing directors, and Greg Sweigart is a senior manager in the Washington National Tax Methods group.

The information in this article is not intended to be “written advice concerning one or more Federal tax matters” subject to the requirements of section 10.37(a)(2) of Treasury Department Circular 230. The information contained herein is of a general nature and based on authorities that are subject to change. Applicability of the information to specific situations should be determined through consultation with your tax adviser. This article represents the views of the authors only, and does not necessarily represent the views or professional advice of KPMG LLP.

Write for Us: Author Guidelines