Paul Starrett
When the question is about technology, the answer has to hold up — technically, legally, and with the people who built it.
I advise and represent companies on tax matters where AI and software are the hard part, and on the AI governance questions that come with them.
Advisory by default, with legal representation where a matter calls for it.
Attorney and technologist. Twenty-five years at the intersection of law and technology, specializing in AI and related technologies since 2013.
Tax — AI and Software Development
The tax answer depends on what the AI actually is and where its value is created.
- When the IRS or FTB examines your credit: audit defense from the first document request, built on what the engineers actually did
- When a prior claim needs a second look: a review of whether it holds up, before an examiner asks, conducted under attorney-client privilege
- When you claim the credit: R&D credit substantiation, federal and California; §174 / §174A
- When you buy or sell: what the AI assets are, and how they’re valued and allocated in an acquisition
- When you sell across states and borders: how AI products and revenue are characterized and sourced
- When IP moves between entities: the technical facts behind where AI value is developed and maintained
I establish which work qualifies and the technical basis for it, including how compute splits between development and production. Your CPA applies that to the numbers, computes the credit, and prepares the return.
AI Governance & Risk
The same analysis, applied to risk: governance has to match what the system actually does, not what the policy says it does.
- When AI makes decisions about people: risk assessment, testing, and documentation before launch, in hiring, lending, health, and housing
- When a customer, investor, or acquirer asks: AI diligence questionnaires and reviews, answered with evidence
- When you rely on vendors’ AI: third-party and AI-vendor risk, privacy, and cybersecurity
- When a regulator or plaintiff asks: regulatory inquiries and defense, backed by the technical record
Disputes & Investigations
Reconstructing what a system did when its outputs come under scrutiny.
Matters like these usually need a technologist, a tax lawyer, and an investigator. You get all three in one engagement, with nothing lost between them, working alongside your CPA, tax counsel, and engineers rather than replacing anyone.
When those roles are split, the gaps get filled with assumptions, and assumptions are exactly what an examiner, a regulator, or opposing counsel will find.
The technology
Understand what the system actually did, how it was built, trained, and run, well enough to answer the second and third questions, not just the first.
- Former security software engineer, RSA Security
- Former General Counsel and Chief Risk Officer, publicly held AI and data-management company
- M.S. Predictive Analytics, Northwestern University
- Lead author, Digital Signatures (RSA Press / McGraw-Hill, 2002)
The law and the tax
Turn the technical facts into legal and tax positions written for the people who will test them: examiners, judges, clients, and co-parties.
- Attorney at Law, California
- Master of Laws in Taxation
The people
Documents never hold every fact. The record has to be confirmed with the engineers, managers, and vendors who built and ran the system, using methods that hold up under scrutiny.
- Certified Fraud Examiner
- Licensed Private Investigator, California
- Author, The Investigative Interviewer's Guidebook, for risk, compliance, and legal professionals (first published 1998)
I have spent my career building technology and practicing law about it: founding chair of the ABA Big Data Committee from 2013 to 2020, more than 45 multimillion-dollar investigation and e-discovery matters for AmLaw 50 firms and Fortune 100 clients, and a contributor to the IAPP's AI Governance certification. In practice since 2001.
I teach as well as practice: AI governance and cybersecurity law at Santa Clara University School of Law, computer and AI forensics in its Graduate School of Engineering, and law and machine learning in the M.S. Data Science program at the University of the Pacific.
Let's discuss your matter.
Advisory by default, with legal representation where a matter calls for it, as counsel or as a licensed investigator under referring counsel’s engagement. The first conversation is direct and confidential.
paul@starrettlaw.com