Science – Apollo Research
https://www.apolloresearch.ai/science
Apollo Research — named in the OpenAI system card for sandbagging and scheming evaluations
2
Lines added
0
Lines removed
Was read
2026-09-20
b584a51a860f8809…
Changed by
2026-09-23
d28e37eddcbae960…
2 lines added
Left column is the version read 2026-09-20, right is 2026-09-23. Unchanged stretches are elided. Both bodies are kept in full and are addressed by the hashes above.
Showing the first 400 lines of the comparison; 56 further lines are held back. This page reports a change, it is not a copy of anyone’s document — the source is linked above and is the authority for what it says.
⋯ 68 unchanged lines
6969 05 July 2026
7070
7171 Science of Scheming
7272
73+evaluations, training run, alignment, scheming, TRE, evals, third party evaluations
74+
7375 Stress Testing Deliberative Alignment for Anti-Scheming Training
7476 Science of Scheming
7577
7678 17 September 2025