Citation·page:safety-spending-at-scale:fn3

Safety Spending at Scale - Footnote 3

Verdictpartial85%

1 check · 4/29/2026

1 → partial

Our claim

entire record

No record data available.

Source evidence

1 src · 1 check

80000hours.org/problem-profiles/artificial-intelligence/resource

partial85%Haiku 4.5 · 4/3/2026

NoteThe source states that there are a few thousand people working on major AI risks, but not all of them are focused on the risks from power-seeking AI. The claim does not make this distinction. The claim states that the number is as of 2025, but the source is a 2023 article that references a 2025 analysis.

Case № page:safety-spending-at-scale:fn3Filed 4/29/2026Confidence 85%