| Before | After | ||
|---|---|---|---|
| 12 | While we do not have definitive evidence that this model could meaningfully help a novice to create severe biological harm—our defined threshold (opens in a new window) for High capability—we have chosen to take a precautionary approach. 2025 System Cards Author OpenAI Keep reading View all Introducing GeneBench-Pro Research Jun 30, 2026 A near-autonomous AI chemist improves a challenging reaction in medicinal chemistry Research Jun 17, 2026 Introducing LifeSciBench Research Jun 17, 2026 | 12 | While we do not have definitive evidence that this model could meaningfully help a novice to create severe biological harm—our defined threshold (opens in a new window) for High capability—we have chosen to take a precautionary approach. 2025 System Cards Author OpenAI Keep reading View all OpenAI Bio Bug Bounty Safety Jul 9, 2026 Separating signal from noise in coding evaluations Research Jul 8, 2026 Introducing GeneBench-Pro Research Jun 30, 2026 |
Follow unlimited companies, monitor the clauses that matter across every platform, and get the full institutional analysis on what each change obligates you to do.