@VitalikButerin: Okay with Kimi 3.1 leaving out top-end offensive cyber capabili

Submit
@VitalikButerin
@VitalikButerin

I am okay with Kimi 3.1 deliberately leaving out top-end offensive cyber capabilities from their training data. Discriminating against use cases is MUCH less bad than discriminating against users. And the world does need all the time it can get to prepare, harden, check old code, (properly) formally verify new code, etc before the full weight of very powerful cyber attacks is upon us. And if this first big wave of AI risk is not handled well, then we may well get a dystopia of highly locked down models, KYC gating on everything, highly intrusive surveillance on everything, a society where you may be a second or third or fourth-class citizen in terms of model access based on your credentials and nationality.

x.com

@VitalikButerin: Okay with Kimi 3.1 leaving out top-end offensive cyber capabili

submitted bytim12,4831h
Read original article
AI summary of the linked article

Vitalik Buterin says he is okay with Kimi 3.1 deliberately leaving top-end offensive cyber capabilities out of its training data. He argues that restricting use cases is much less harmful than restricting users, and that the world needs time to harden systems, check old code, and formally verify new code before powerful cyberattacks arrive. He warns that if this first wave of AI risk is handled poorly, it could lead to a dystopia of locked-down models, KYC gating on everything, intrusive surveillance, and citizens ranked by model access based on credentials and nationality.

Comments
No comments yetWhat caught your eye? Start the discussion.
Top stories that week →