Are you thinking about model capabilities in coding and math starting late 2025 or so? Those were intentionally boosted via automated RLVR, and there's nothing even loosely comparable to that in applied biology work, let alone in the speculative "helping a bad actor do something crazy" domain that the AI safety folks are worried about. You can't extrapolate from one to the other.
The implied concerns from sensible safety advocates are also about someone jailbreaking the latest proprietary AI frontier model for something like this (which is why their current guardrails are so extreme), not about toy local models.
alt Hacker News
Are you thinking about model capabilities in coding and math starting late 2025 or so? Those were intentionally boosted via automated RLVR, and there's nothing even loosely comparable to that in applied biology work, let alone in the speculative "helping a bad actor do something crazy" domain that the AI safety folks are worried about. You can't extrapolate from one to the other.
The implied concerns from sensible safety advocates are also about someone jailbreaking the latest proprietary AI frontier model for something like this (which is why their current guardrails are so extreme), not about toy local models.