AGI Strategies

person

Max Bartolo

Cohere; LLM evaluation researcher

Researcher at Cohere; previously a UCL DeepMind PhD student. Co-developed adversarial-training and evaluation methods for question-answering and instruction-following.

current Research Scientist, Cohere

Strategy positions

Evals-drivenendorses

Capability/risk evals gate deployment; evals are the load-bearing artefact

Argues evaluation methods that adversarially probe model weaknesses are the only way to characterize what models will do in deployment; static benchmarks are insufficient.

Adversarial evaluation reveals failure modes that static benchmarks miss. As models become more capable, our evaluation has to become more adversarial too.
articleMax Bartolo, research page· maxbartolo.com· 2024· faithful paraphrase

Closest strategy neighbours

by jaccard overlap

Other people whose strategy tags overlap with Max Bartolo's. Overlap is on tag identity, not stance; opposites can show up if they reference the same tags.

  • Aleksander Mądry

    shared 1 · J=1.00

    MIT; ex-OpenAI head of preparedness

  • Alex Meinke

    Alex Meinke

    shared 1 · J=1.00

    Apollo Research; deceptive alignment evaluations

  • Ali Rahimi

    Ali Rahimi

    shared 1 · J=1.00

    Google Brain ML researcher; 'Alchemy' speech

  • Anna Rogers

    Anna Rogers

    shared 1 · J=1.00

    IT University of Copenhagen; LLM benchmarking critique

  • Arati Prabhakar

    Arati Prabhakar

    shared 1 · J=1.00

    White House OSTP director (2022–2025)

  • Beth Barnes

    Beth Barnes

    shared 1 · J=1.00

    Founder of METR; dangerous capability evaluations

Record last updated 2026-04-25.