person
Hjalmar Wijk
METR researcher; AI R&D evaluations
Researcher at METR; co-author of evaluation studies on whether frontier models can autonomously do AI R&D tasks at the level of professional researchers.
current Researcher, METR
Strategy positions
Evals-drivenendorses
Capability/risk evals gate deployment; evals are the load-bearing artefactArgues standardized AI-R&D benchmarks, where models are evaluated on the very work that would fuel recursive self-improvement, are an important safety signal we currently lack.
We measure how well frontier models can perform AI R&D tasks compared to human researchers. The gap is closing in some specific dimensions and that is what an early-warning system should be tracking.
Closest strategy neighbours
by jaccard overlapOther people whose strategy tags overlap with Hjalmar Wijk's. Overlap is on tag identity, not stance; opposites can show up if they reference the same tags.
Record last updated 2026-04-25.