person
Stuart Armstrong
Aligned AI co-founder; ex-FHI; value-extrapolation approach
Philosopher and AI safety researcher who spent over a decade at the Future of Humanity Institute. Co-founded Aligned AI; his research centres on value extrapolation, the hypothesis that solving how to extend human values across contexts is necessary and nearly sufficient for alignment.
Profile
expertise
Deep technical
Sustained peer-reviewed contribution to ML, alignment, interpretability, or safety techniques. Could review a frontier paper.
Founded Aligned AI (now Aligned). Long FHI safety publication record on value learning, corrigibility, indifference.
recognition
Established
Reliable, recognised voice within their specific subfield. Cited and invited but not central to general AI discourse.
Recognised within alignment field; little mainstream profile.
vintage
Pre-deep-learning
Active before AlexNet. The existential-risk frame matures (FHI, OpenPhil, EA). Public AI commentary still rare; deep learning not yet dominant.
FHI from 2010. Long body of agent-foundations safety work pre-AlexNet; Aligned AI is a continuation of that frame.
Hand-classified. See the board for the criteria and the full grid.
Strategy positions
Alignment firstendorses
Solve technical alignment before capability thresholds closeArgues alignment is solvable through value-extrapolation techniques; publicly optimistic about the tractability of the problem after a decade of theoretical AI safety research.
“The challenge of getting AIs to follow human values not only must be solved, but can be solved, and will be solved.”
Humans are weak agents in a strong sense: we can describe the world, but not always our values. Alignment has to work with that.
Closest strategy neighbours
by jaccard overlapOther people whose strategy tags overlap with Stuart Armstrong's. Overlap is on tag identity, not stance; opposites can show up if they reference the same tags.
Record last updated 2026-04-25.