For information only. Not advice, and not to be relied on. What that means, in full

Science – Apollo Research

https://www.apolloresearch.ai/science

Apollo Research — named in the OpenAI system card for sandbagging and scheming evaluations
1
Lines added
1
Lines removed
Was read
2026-09-11
116cd2f745bf013d
Changed by
2026-09-20
b584a51a860f8809

1 line added, 1 line removed

Left column is the version read 2026-09-11, right is 2026-09-20. 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; 55 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.
61 unchanged lines
6262 21 July 2026
6363  
6464 Science of Scheming
6565  
66We need 3rd party Training-Run Assessments
66+We need 3rd party Training-Run Evaluations
6767 Science of Scheming
6868  
6969 05 July 2026
7070  
GovernanceHub — the governance registry and policy router for AI systems