Join our team
Work with us directly to advance AI safety research. No roles are open at the moment. If you think you may be a good fit for the team, introduce yourself and we will reach out when something opens up.
Research
A selection of research conducted by Pivotal fellows.
35 research results
Paper · 12 Oct 2026
Audit tools of the trade: Automated Auditor Elicitation Can Be Sensitive to the Context the Auditor Is Given
Researchers: Ben Slater, Ryan Lundqvist, Leo Richter & Kevin Wei
Mentor: Kevin Wei
Technical AI Safety
Paper · 31 Aug 2026
Workload Identification with Physical Side Channels for AI Governance
Researcher: Simone Gargiulo
Mentor: Gabriel Kulp
Technical AI Governance
Paper · 12 Aug 2026
Silent Updates: Measuring and Closing the Post-Deployment Disclosure Gap
Researcher: Sophia Abraham
Mentor: Ben Bucknall
AIES 2026
Technical AI Governance
Paper · 3 Aug 2026
Privacy-Preserving AI Verification via Minimal Information Disclosure
Researcher: Sleem Abdelghafar
Mentor: Gabriel Kulp
arXiv preprint
Technical AI Governance
Paper · 1 Aug 2026
A False Average: Chain-of-Thought Monitors Collapse Where They Are the Only Defense
Researcher: Shikhar Shiromani
Mentor: Noah Y. Siegel
arXiv preprint
Technical AI Safety
Blog post · 8 Jul 2026
Can the U.S. and China Deny AI?
Researcher: Felix Choussat
Mentor: Oscar Delaney
Policy & Governance
Paper · 6 Jul 2026
Compressed Computation under L⁴ Loss is likely Computation in Superposition
Researcher: Francisco Ferreira da Silva
Mentor: Stefan Heimersheim
arXiv preprint
Technical AI Safety
Paper · 3 Jul 2026
Individual Parameters in Weight-Sparse Transformers Appear Interpretable
Researcher: Arnau Marin-Llobet
Mentor: Stefan Heimersheim
ICML 2026: Mechanistic Interpretability Workshop
Technical AI Safety
Blog post · 24 Jun 2026
Door's Locked, Try the Window: Measuring Circumvention Propensity in Coding Agents
Researcher: Prakrat Agrawal
Mentor: Jérémy Scheurer
Technical AI Safety
Paper · 23 Jun 2026
Evidence for Feature-Specific Error Correction in LLMs
Researcher: Francisco Ferreira da Silva
Mentor: Stefan Heimersheim
arXiv preprint
Technical AI Safety
Paper · 1 Jun 2026
Denying Personhood to AI: An Analysis of U.S. State Legislation on AI Legal Status
Researcher: Austin Smith
Mentor: Lucius Caviola
Policy & Governance
Paper · 14 May 2026
Training on Documents About Monitoring Leads to CoT Obfuscation
Researcher: Reilly Haskins
Mentors: Joshua Engels & Bilal Chughtai
arXiv preprint
Technical AI Safety
Paper · 3 May 2026
GPU Fingerprinting for Location Verification
Researcher: Wayne Tee
Mentor: Jonathan Happel
ICML 2026: Workshop on Technical AI Governance
Technical AI Governance
Paper · 30 Apr 2026
Causal Foundations of Collective Agency
Researcher: Frederik Hytting Jørgensen
Mentor: Lewis Hammond
Proceedings of CLeaR 2026 (PMLR 323)
Technical AI Safety
Blog post · 28 Apr 2026
AI Access Is Not Enough: Middle Powers Need Strategic Reserves
Researcher: Kasia Jakimowicz
Mentors: Prof. Robert Trager & Charles Martinet
Policy & Governance
Paper · 23 Apr 2026
Bayesian Influence Functions for Hessian-Free Data Attribution
Researcher: Philipp Alexander Kreer
Mentor: Jesse Hoogland
ICLR 2026 Poster
Technical AI Safety
Paper · 7 Apr 2026
Stochastic Gradient Descent in the Saddle-to-Saddle Regime of Deep Linear Networks
Researchers: Guillaume Corlouer & Avi Semler
Mentors: Alexander Strang & Alexander Gietelink Oldenziel
arXiv preprint
Technical AI Safety
Paper · 4 Feb 2026
Attack Selection Reduces Safety in Concentrated AI Control Settings against Trusted Monitoring
Researchers: Joachim Schaeffer & Arjun Khandelwal
Mentor: Tyler Tracy
ICLR 2026: AIWILD Workshop
Technical AI Safety
Paper · 1 Dec 2025
Factor(T,U): Factored Cognition Strengthens Monitoring of Untrusted AI
Researcher: Aaron Sandoval
Mentor: Cody Rushing
arXiv preprint
Technical AI Safety
Paper · 1 Dec 2025
Beyond Vibe Decision Theory: Asymmetric Manipulation Vulnerabilities in LLM Multi-Agent Coordination
Researcher: Sukanya Krishna
Mentor: Tobin South
NeurIPS 2025 Workshop on Algorithmic Collective Action; AAMAS 2026
Technical AI Safety
Paper · 20 Nov 2025
The Multi-Agent Off-Switch Game
Researcher: Soroush Ebadian
Mentor: Lewis Hammond
Trustworthy Agentic AI Workshop@AAAI26
Technical AI Safety
Paper · 12 Nov 2025
Decomposition of Small Transformer Models
Researcher: Casper L. Christensen
Mentor: Logan Riggs Smith
NeurIPS 2025: Mechanistic Interpretability Workshop
Technical AI Safety
Paper · 14 Oct 2025
Influence Dynamics and Stagewise Data Attribution
Researchers: Jin Hwa Lee & Matt Smith
Mentor: Jesse Hoogland
ICLR 2026
Technical AI Safety
Paper · 30 Sep 2025
Activation Probes Are Reliable With a Handful of Positive Examples
Researcher: Riya Tyagi
Mentor: Stefan Heimersheim
NeurIPS 2025: Mechanistic Interpretability Workshop
Technical AI Safety
Paper · 26 Jul 2025
Trivial Trojans: How Minimal MCP Servers Enable Cross-Tool Exfiltration of Sensitive Data
Researcher: Nicola Croce
Mentor: Tobin South
Technical AI Governance Forum 2025
Technical AI Safety
Paper · 16 Jul 2025
Benchmarking Deception Probes via Black-to-White Performance Boosts
Researcher: Avi Parrack
Mentor: Stefan Heimersheim
ICML 2025: Actionable Interpretability Workshop
Technical AI Safety
Report · 17 Jun 2025
Safety Features for a Centralised AGI Project
Researcher: Sarah Hastings-Woodhouse
Mentor: Elliot Jones
arXiv preprint
Policy & Governance
Paper · 26 May 2025
Understanding the learned look-ahead behavior of chess neural networks
Researcher: Diogo Cruz
TMLR
Technical AI Safety
Paper · 14 Apr 2025
Product Liability as a Model for UK AI Security
Researcher: Gaurav Yadav
Mentor: Peter Wills
Policy & Governance
Blog post · 28 Mar 2025
Forging the Biological Weapon Convention: A Brief History of the Creation of the BWC
Researcher: Neha Suresh
Mentors: Alex DeMarsh & Gabby Essix
The Pandora Report
Biodefense
Blog post · 4 Dec 2024
Will the US Government Control the First AGI?—Finding Base Rates
Researcher: Luise Woehlke
Mentor: Christian Ruhl
Policy & Governance
Blog post · 4 Dec 2024
Sharing the AI Windfall: A Strategic Approach to International Benefit-Sharing
Researcher: Michel Justen
Mentors: Matthew van der Merwe & Max Dalton
Policy & Governance
Paper · 7 Nov 2024
The Role of AI Safety Institutes in Contributing to International Standards for Frontier AI Safety
Researcher: Kristina Fort
Mentor: Oliver Guest
arXiv preprint
Policy & Governance
Paper · 3 Jun 2024
On Labs and Fabs: Mapping How Alliances, Acquisitions, and Antitrust are Shaping the Frontier AI Industry
Researcher: Tomás Aguirre
arXiv preprint
Policy & Governance
Paper · 14 Apr 2023
How to design an AI ethics board
Researcher: Anka Reuel
Mentor: Jonas Freund
AI and Ethics
Policy & Governance
Get involved
Mentor top emerging researchers
Guide a fellow on a research project, share your expertise, and help them grow as a researcher. We handle recruitment, logistics, and day-to-day research support.
Participate in a fellowship
Join our in-person research fellowship in London and work with experienced mentors on AI safety, AI governance, AIxBio, or biodefense.