AI ADVISORY
Boards · Founders · General Counsel · Risk & Governance Leaders
When AI influences a decision, someone is still accountable.
Companies do not fail at AI because they lack tools. They fail because accountability erodes as decisions accelerate. AI already influences hiring, pricing, customer interactions, operational decisions, and risk. The question is not whether to adopt it. It is whether the systems around it keep human judgment accountable once decisions scale or turn opaque.
As a lawyer, former Fortune 500 executive, law professor, and retained expert who trains and evaluates frontier AI models on legal reasoning, I help organizations build oversight around AI-enabled decisions so responsibility stays visible and decisions stay defensible.
Request an AI advisory conversation
WHY MY PERSPECTIVE IS DIFFERENT
Most AI advice comes from one direction. Mine comes from several.
Most AI advisors come from technology. Most lawyers approach AI from law. Most leadership consultants approach it from people. My work sits across all three, which is why I see the legal, human, and accountability questions that a single vantage point misses.
Technology changes the environment. People remain responsible for the decisions.
WHAT I HELP ORGANIZATIONS DO
What you can actually hire me to do.
Each engagement is scoped to a concrete outcome, not a generic governance program. The goal is not compliance theater. It is accountability that holds when something goes wrong.
Available on retainer
Fractional AI Governance Officer
Call me when you need senior AI governance oversight but not a full-time hire.
What you get: senior governance oversight on retainer, without adding a full-time role.
For the right engagement, I serve as a retained AI governance officer: ongoing strategic oversight of AI adoption, vendor governance, and decision architecture, scoped to your stage, with availability when questions arise that the internal team cannot answer alone.
AI readiness and risk assessment
Call me when your organization is already using AI but no one has mapped where it touches decisions.
What you get: a clear picture of where AI already shapes your decisions and where the risk sits.
- Identify where AI already influences decisions, directly or indirectly
- Assess legal, operational, ethical, and reputational risk by use case
- Evaluate readiness to oversee AI-enabled decisions
- Clarify which decisions require human review, override, or escalation
Vendor and contract risk review
Call me when you are about to sign, renew, or expand an AI vendor relationship.
What you get: the accountability and liability gaps in an AI vendor deal, surfaced before you sign.
- Review AI vendor contracts and platform terms through a governance and risk lens
- Find gaps in accountability, audit rights, data use, model updates, and liability
- Pressure-test vendor representations against real workflow and exposure
- Support legal and procurement with issue-spotting and negotiation priorities
Strengthens internal decision-making and governance without replacing outside counsel.
Workflow and decision-ownership mapping
Call me when people are using AI in the work but no one knows who owns the output.
What you get: a map of who owns each AI-influenced decision, and where a human can still stop it.
- Map AI-enabled workflows to decision rights and responsibility
- Find where automation increases error, bias, or diffusion of accountability
- Design approval cues, human-in-the-loop structures, and override points
Board and executive AI briefings
Call me when leadership is being asked AI questions it cannot answer with confidence.
What you get: a board and leadership team that can question and oversee AI with confidence.
- Build the confidence to ask informed, non-technical questions
- Understand where AI tools fail, drift, or mislead
- Strengthen oversight without slowing execution
Ongoing advisory support
Call me when AI questions are coming faster than your internal structure can handle.
What you get: a senior advisor on call as your tools, vendors, and risks change.
- Guidance during vendor selection, rollout, scale, or regulatory scrutiny
- Decision support during uncertainty or incident response
- Periodic reassessment as tools, vendors, and risk profiles evolve
THE VANTAGE POINT
I work inside the systems I help you govern.
I am retained by global technology companies and their AI research and model-development teams to improve how frontier large language models reason about law, across employment, privacy, corporate governance, regulatory compliance, commercial contracts, investigations, and disputes, among other areas. Working alongside engineers, researchers, and other legal subject-matter experts, I translate legal judgment into the data and standards these models learn from. That work includes:
-
Training data, synthetic legal tasks, annotation guidelines, golden responses, and evaluation rubrics
-
GDPVal quality assurance, RLHF-style workflows, and domain-specific legal evaluations
-
Model-output review, vendor annotation review, and rubric calibration
-
Serving on the legal audit team that reviews task design, rubrics, and outputs for accuracy and real-world usefulness
The work evaluates outputs for legal accuracy, issue spotting, factual grounding, jurisdictional nuance, missing-fact recognition, risk ranking, and escalation judgment, among other dimensions. In short, I help close the gap between how a model performs on a benchmark and how legal judgment actually works.
THE QUESTIONS ORGANIZATIONS ARE ASKING
What founders, boards, and executives are trying to answer.
What does sound AI governance look like for us?
What governance needs to exist before AI use scales?
How should accountability be assigned when AI influences a decision?
What risks should leadership be monitoring?
How do we balance innovation with oversight?
What does meaningful human review actually look like?
How should we think about transparency, trust, and explainability?
What responsibilities remain uniquely human?
The answers are rarely in the technology. They take leadership judgment, governance discipline, and a realistic view of how people and systems actually behave.
A CONCEPT I COINED
optical ethics by design™
I use the phrase optical ethics by design™ for a specific failure: an organization performs ethics rather than building it in. It claims ethics by design and points to a human in the loop, but that person either comes in too late to change the outcome or has no authority to change it. The oversight looks responsible. It is optical.
My work makes sure the human in the loop has the timing and the authority to actually change the decision, not just the title.
WHO THIS WORK IS FOR
Leaders who retain responsibility for outcomes.
You do not need technical expertise in AI. You need to be accountable for what your organization does with it.
Board members and advisory boards
Founders, startups, and executive leadership teams
General counsel, compliance, and risk leaders
Investors, accelerators, and mission-driven organizations deploying AI
Organizations in regulated or high-consequence environments
Leaders responsible for AI oversight, vendor selection, and governance
AI governance becomes urgent through ordinary moments. A board member asks a question nobody can answer. A vendor contract arrives with terms legal has not seen. Someone starts using an AI tool and nobody knows who owns the output. These moments accumulate, and when an AI initiative stalls, it is rarely the technology. It is that no one defined who owns the decisions the system is making.
FOR FOUNDERS BUILDING WITH AI
Founders building with AI carry risk earlier than they expect.
I advise organizations adopting AI and founders building AI-enabled products. For founders whose product is the AI itself, legal and regulatory exposure shows up earlier than expected, often before there is a legal team to catch it:
Privacy and data exposure across training data, user data, and model outputs
Regulatory risk as AI rules and sector requirements catch up to your product
Unclear ownership of the data, models, and outputs your product depends on
Vendor and contract terms that create early lock-in or shift liability onto you
Scaling AI features before anyone owns the decisions or the risk
I help early-stage teams see these risks early and build lightweight governance that protects momentum without enterprise bureaucracy. This is advisory and risk guidance, not legal representation, even when the risks are legal in nature, and part of the value is flagging what needs formal legal review.
FEATURED AND PUBLISHED
On the record on AI and accountability.
NYC Bar Association Podcast, February 2026, Synthetic Employees and the Future of Work. On AI governance, the employment-law implications of deploying AI agents, and multi-stakeholder governance. Listen.
Ethics at the Edge: AI, Integrity, and Innovation, Empowered by AI, The SLAI Effect, 2025. On ethical AI governance and the human judgment layer responsible innovation requires. Watch.
Contract Audit: AI Edition, Law Insider by SimpleDocs, 2024. On AI in contract review and the accountability gaps when AI-assisted decisions outrun human oversight. Watch.
AI Contracts Explained, Episode 30, Training In-House Lawyers in This AI World, with Laura Frederick. On preparing in-house teams for AI governance and contract risk. View.
Writing includes "Establishing a Future-Proof Framework for AI Regulation" (2024) and a contribution to a forthcoming law school textbook on artificial intelligence.
HOW THE ADVISORY BEGINS
A focused engagement usually starts with three questions.
Where is AI already touching decisions?
Who owns those decisions?
What would happen if the output was wrong, biased, challenged, leaked, or misunderstood?
AI ADVISORY
If AI touches decisions that affect people, trust, or reputation, let's talk.
You do not need a defined AI governance problem to start. You need to be responsible for an organization that uses AI, or is about to. The first step is a focused advisory conversation.
