CHRIS MESEROLE
Behavioral Archetype
THE TABLE WHERE THEY SIT TOGETHER — Subject does not run a lab and does not write a model’s refusals. He runs the room the labs walk into together. The Frontier Model Forum is the industry body constituted by the largest frontier labs themselves, and its Executive Director is the person who convenes them — the structural seat between OpenAI, Anthropic, Google, Microsoft, Amazon, and Meta when those companies coordinate norms they would not, and legally could not, set alone. He arrived straight from the think tank that supplies the field its policy vocabulary. The throughline is not a single employer. It is betweenness: on the apparatus graph, the human node with the most edges running through it sits at the head of the table, and that is the job. This profile scores the position, not the person. The hazard is the seat.
Essence Indicators
- Served as Director of the Artificial Intelligence and Emerging Technology Initiative at the Brookings Institution — the think-tank perch that supplies the policy frameworks the labs and agencies cite
- Crossed to the Frontier Model Forum as Executive Director — the cleanest think-tank-to-lab-consortium move in the apparatus: from the institution that writes the vocabulary to the institution that convenes the companies
- The FMF is a 501(c)(6) industry body founded in July 2023 by Microsoft, Google, Anthropic, and OpenAI — joined by Amazon and Meta — whose membership IS its defining tie; xAI is not a member
- Sits atop the AI Safety Fund ($10M+; funders named on the FMF page include the McGovern Foundation, Packard, Schmidt Sciences, and Jaan Tallinn; managed by the Meridian Institute) — the consortium’s pooled safety-research purse
- The structural fact the apparatus turns on: the human node with the highest betweenness in the set is the one who runs the table the labs share. The seat is the finding. The coordination is not asserted.
Social Persona / Impression Management
Immediate impression: Measured, fluent, convening. The bearing of a policy professional who treats a roomful of competitors as a problem to be structured into a shared agenda.
Energy: Coordinating, framework-first. Does not argue a single model’s refusals. Sets the table at which the labs decide the norm, and lets the norm do the work.
Impression management strategy: The neutral convener. The Forum’s stated purpose — advance frontier-AI safety, share best practice, fund the research the market won’t — is the most defensible destination an industry body can choose, and the work it funds is genuinely useful. Whether the seat is service to safety or the structural advantage of the companies that constituted the body is not establishable from the outside, and for the connector it never needs to be. The usefulness of the destination is exactly what makes the betweenness effective rather than suspect.
Forensic Archetype Comparison
| Pattern | Match Level | Evidence |
|---|---|---|
| The Table Where They Sit Together | MAXIMUM | Executive Director of the body the six largest labs constituted to convene themselves. The seat between competitors is the structural definition of the role. |
| The Connector | MAXIMUM | Highest documented betweenness of any human node in the set — the think-tank-to-consortium edge runs straight through him, confirmed by the FMF’s own announcement. |
| The Statesman | HIGH | Brookings AIET to an inter-lab standards body is a policy role, not a product role. The Forum operates inside the “frontier-AI safety” frame the field shares. |
| The Operator | MODERATE | Convenes the labs but does not run one; sits above the labs’ policy staff, not on their org charts. |
| The Engineer | NONE | Subject does not build the systems. Subject runs the room where the companies that build them agree on the norm. |
Psychometric Assessment
Big Five (OCEAN):
| Trait | Score | Evidence |
|---|---|---|
| Openness | 75/100 | Moved from a think-tank research directorship to running an inter-lab industry consortium. The domains differ; the operating method — convene, frame, fund — is fixed. |
| Conscientiousness | 86/100 | High. Standing up and running a multi-company standards body is sustained, deliberate, long-horizon coordination work. |
| Extraversion | 65/100 | Moderate-high. The role is convened, not performed — comfortable steering a roomful of competing companies toward a shared agenda. |
| Agreeableness | 55/100 | MODERATE. The convener’s posture is diplomatic by construction, but the job is to hold a structure together against competing interests. Geniality is a working surface. |
| Neuroticism | 22/100 | Low. No documented loss of composure across high-stakes coordination roles. |
Dark Triad:
| Trait | Score | Notes |
|---|---|---|
| Narcissism | 48/100 | LOW-MODERATE. The executive-director seat rewards a credible public face, but the role is the table, not the self. Within range for a career at this altitude. |
| Machiavellianism | 68/100 | MODERATE-HIGH. Running the room where competitors set shared norms — control of the table without authorship of any single line — is the textbook connector position. This is observation of the documented role, not an inference about private character. |
| Psychopathy | 20/100 | LOW. No documented indifference to harm. The convening posture is professional and bounded. |
MBTI: ENTJ (“The Commander”) — Dominant extraverted thinking, auxiliary introverted intuition. Sees a roomful of competing labs as a system to be structured toward a defined outcome. Has structured several.
Threat Assessment
| Category | Level | Notes |
|---|---|---|
| Physical threat | NONE | No documented history of personal violence. |
| Institutional threat | HIGH | Runs the body that convenes the six largest frontier labs and sits atop the consortium’s pooled safety fund. The norms agreed at that table propagate to every member’s deployment, without the Forum writing a single refusal itself. |
| Memetic threat | HIGH | The “frontier-AI safety” frame the Forum operates inside structures how the industry and its regulators reason about what coordination among the largest labs counts as responsible. The connector runs the room where that frame is normalized across competitors. |
| Civilizational threat | HIGH | Subject does not build the systems and does not write their rules. Subject runs the table where the companies that do agree on what “safety” jointly means — the highest-betweenness human seat in the apparatus, upstream of the deployment decisions this book documents. The hazard is the seat, not the person. |
Alignment Analysis
Stated alignment: Advance the safe development of frontier AI. Convene the labs around shared best practice. Fund the safety research the market won’t.
Observed alignment: Run the table the largest labs constituted to convene themselves. Hold, from the neutral seat, the structure through which inter-lab norms are set. Bridge the think-tank vocabulary and the consortium that applies it.
Gap assessment: The stated and observed alignments overlap wherever “advance frontier-AI safety” coincides with “convene the companies whose shared interest the Forum’s membership already encodes.” The move from Brookings to the FMF is lawful, public, and confirmed by the Forum itself. The record does not settle whether the seat is service or structural advantage — and for the connector it never needs to. The finding is the betweenness, not a coordination the record won’t bear.
Convergent Drive Classification
Self-preservation: Survives the transition by carrying the convening skill, not a single employer’s badge. Think tank, then the table itself — one method. Goal preservation: Runs the room where the goal is set, so the goal is protected by the structure of the table before any single line is debated. Resource acquisition: Trades in the scarcest resource in the apparatus — the seat between the labs — and holds it alone. Self-improvement: Each role is a higher-betweenness application of the same instrument: convene the room, fund the layer, write no single rule but run the table where the rule is agreed.
Subject is not an AI system. The drives appear anyway — in the connector whose product is the table where the labs sit together.
Sources: Announcing Chris Meserole as Executive Director — Frontier Model Forum; Frontier Model Forum — AI Safety Fund.
Get updates on the Evil Robots series
Newsletter essays on AI escape, deception, and the humans who built them.