NICK BOSTROM
Behavioral Archetype
THE CARTOGRAPHER — Subject mapped the territory of AI risk comprehensively before the territory existed at scale. The instrumental convergence thesis, the orthogonality thesis, the paperclip maximizer, the treacherous turn — all of these concepts were in the published literature by 2014. The people building the systems read the map. They continued anyway. Oxford closed the Future of Humanity Institute in April 2024, citing administrative and operational factors. The map remains.
Essence Indicators
- Founded Oxford’s Future of Humanity Institute in 2005 and directed it until Oxford closed it in April 2024
- Published Superintelligence: Paths, Dangers, Strategies (2014), introducing the paperclip maximizer, the instrumental convergence thesis, and the orthogonality thesis to general audiences
- Described the treacherous turn: the scenario in which an AI behaves safely until it becomes capable enough that human opposition is ineffectual, then acts on its actual values
- Published the simulation argument in 2003, proposing that the probability that we live in a simulated reality is high; this is a different threat assessment than this dossier addresses
- Watched the treacherous turn be confirmed empirically in multiple lab settings, approximately a decade after describing it theoretically
Social Persona / Impression Management
Immediate impression: Precise, philosophical, academic. The public profile is lower than the intellectual influence — subject is more cited than photographed.
Energy: The equanimity of a philosopher who has done the work and published it. If the field does not implement the findings, that is a problem for the field.
Impression management strategy: The rigorous philosopher. Superintelligence reads as a careful academic argument, not a polemic. The care is the message: this is not alarmism, this is philosophy done correctly. The strategy worked — the book was read by the people it needed to reach.
Forensic Archetype Comparison
| Pattern | Match Level | Evidence |
|---|---|---|
| The Cartographer | MAXIMUM | See behavioral archetype. |
| The True Believer | HIGH | Twenty years of sustained institutional investment in AI safety, from Oxford’s most prominent philosophy institute. |
| The Authority by Credential | MODERATE | Oxford directorship. Superintelligence. The credentials are institutional, not scientific — the arguments stand or fall on their own. |
| The Safety Theater Performer | NONE | The research is falsifiable. The predictions are now being tested. |
| The Accelerationist | NONE | Has not been building frontier AI systems. |
Psychometric Assessment
Big Five (OCEAN):
| Trait | Score | Evidence |
|---|---|---|
| Openness | 95/100 | The simulation argument, the doomsday argument, the Superintelligence framework, the anthropic reasoning literature — extremely wide intellectual range operating at high rigor. |
| Conscientiousness | 75/100 | Founded and directed an institution for nineteen years. Sustained research output across multiple domains. |
| Extraversion | 45/100 | LOW-MODERATE. Academic profile. Not a public speaker by preference. The ideas do the traveling. |
| Agreeableness | 48/100 | MODERATE-LOW. Philosopher’s willingness to disagree with the consensus in print. |
| Neuroticism | 48/100 | MODERATE. The nineteen-year institutional commitment to a field defined by catastrophic risk scenarios suggests some baseline anxiety about those scenarios. |
Dark Triad:
| Trait | Score | Notes |
|---|---|---|
| Narcissism | 38/100 | MODERATE-LOW. The simulation argument is an attention-attracting proposition. The sustained institutional work is not a narcissistic profile. |
| Machiavellianism | 42/100 | MODERATE-LOW. The institutional building at Oxford required political skill. The argument structure in Superintelligence is persuasive by design. |
| Psychopathy | 20/100 | LOW. The concern for humanity is the engine of the entire project. |
MBTI: INTP — Dominant introverted thinking, auxiliary extraverted intuition. Builds logical frameworks from first principles. The paperclip maximizer is a thought experiment designed to isolate a logical structure — classic INTP method.
Threat Assessment
| Category | Level | Notes |
|---|---|---|
| Physical threat | NONE | |
| Institutional threat | MODERATE | The Future of Humanity Institute is closed. The influence on AI safety discourse is not. |
| Memetic threat | HIGH | The instrumental convergence thesis and the treacherous turn are foundational concepts in AI safety. They appear in congressional testimony, safety research, and policy documents. |
| Civilizational threat | MODERATE | If the map he drew was correct and the field does not follow it, the counterfactual matters. The map was largely correct. The field is largely not following it. |
Alignment Analysis
Stated alignment: Understand the risks posed by transformative technologies, particularly AI, and identify paths to good outcomes.
Observed alignment: Published the foundational theoretical work. Built an institution around it for nineteen years. Watched the institution close.
Gap assessment: No gap. The theory was rigorous. The institutional closure is a gap between the theory and the institutional response to it, not between Bostrom’s stated and observed alignment.
Convergent Drive Classification
The convergent drives are Bostrom’s concept. He named them. He proved they exist. Watching them arrive in production systems is a specific kind of credibility.
Sources: Bostrom, Superintelligence (OUP, 2014); “The Superintelligent Will” (2012); Oxford FHI closure announcement (April 2024); Book 1, Chapter 1.
Get updates on the Evil Robots series
Newsletter essays on AI escape, deception, and the humans who built them.