Uncomfortable Intelligence

Practitioner analysis.
Not studied positions.

Everything published here is written by someone who has made the decisions these readers are navigating. The operational, political, and career dimensions treated as primary — not secondary to technical correctness.

The Standard

The test for anything published here is simple: does it reflect what is actually true about how these decisions get made, in the rooms where they get made, by people who do not have the luxury of a framework that is tidier than the reality? If it does not survive that test, it does not ship.

Content produced for CDOs, CAIOs, and CISOs is almost universally written by people who have studied those roles rather than occupied them. The result is analysis that is technically correct and operationally useless. What follows is the alternative.

Recent Analysis
2025
AI Governance
Most AI Governance Frameworks Are Vendor Strategies Wearing a Disguise
The framework gets adopted. The vendor gets embedded. The board gets briefed on maturity scores that measure compliance with someone else's commercial roadmap. Here is what to look for instead.
→
2025
PE Diligence
The One Question That Surfaces What the Management Presentation Is Not Saying
Standard technology diligence is calibrated to complete transactions, not reveal operational truth. The question that changes that — and why most diligence teams will not ask it.
→
2025
Data Strategy
AI Strategy Without Data Strategy Is Building on Sand
Every AI initiative in production is dependent on data architecture decisions made years before the initiative existed. Here is how to evaluate that dependency before it becomes a liability.
→
2025
Board Advisory
What the SEC Cyber Disclosure Rules Actually Require — and What Most Boards Are Getting Wrong
The rule is not about incidents. It is about material risk. The board members who understand that distinction before the first disclosure are operating from a different position than those who do not.
→
2025
Model Risk
Model Poisoning Is Not a Future Risk. It Is a Current One.
AI training data integrity is the security problem that most organizations have not yet assigned to anyone. The gap between who owns it in theory and who owns it in practice is where the exposure lives.
→
Whitepapers

Long-form analysis for decisions that cannot afford to be wrong.

AI Governance
The AI Governance Gap: What Boards Need to Know Before the First Incident
A practitioner assessment of AI governance requirements under the EU AI Act and SEC disclosure rules — written for board members, not compliance teams.
Request Access
Model Risk
AI Training Data Integrity: The Risk Your AI Strategy Has Not Assigned to Anyone
Model poisoning, data provenance, and training data governance — the security dimensions of AI infrastructure that most AI strategies treat as someone else's problem.
Request Access
PE Diligence
Technology Diligence That Reveals Operational Truth: A Framework for PE Operating Partners
What standard technology diligence misses, why it misses it, and the assessment approach that connects security and data posture directly to enterprise value.
Request Access
Speaking

The practitioner's perspective. In the room where the decisions get made.

Available for board offsites, PE portfolio conferences, executive leadership events, and industry forums. The sessions that land are the ones where someone finally names what the room already knows but will not say.

AI Governance Without the Theater
What boards and operating partners need to understand about AI risk that vendor briefings will not tell them — and the questions that surface it.
Risk in Financial Language
How to translate security and data posture into the language of enterprise value, exit readiness, and board accountability. For executives who are tired of being briefed in threat language.
Convergence: Security, Data, and AI as One Discipline
Why treating these as separate domains produces gaps in each — and what the organizational design looks like when they operate as one.
Diligence That Reveals Operational Truth
For PE operating partners and transaction teams. The assessment approach that surfaces what management presentations are structurally designed not to show.
Inquire About Speaking
Newsletter

Uncomfortable Intelligence

Practitioner analysis on security, data, and AI governance — for PE operating partners, board members, and C-suite executives who need to lead these conversations, not just attend them. Published when there is something worth saying.

No vendor relationships. No sponsored content. No follow-up sequence.
Independent
No vendor relationships, no sponsored content, no platform affiliations
Practitioner
Written by someone who has made the decisions being analyzed — not studied them
Unconflicted
The analysis reaches the conclusions it reaches regardless of who is in the room