Thoughts While Watching AI Regulation Discussions
These days, there's a lot of talk in various countries about AI regulation bills, but it feels like the regulators don't understand the technology well, and the opponents only talk about their own interests, which is a bit frustrating.
Of course, deepfakes and copyright issues are really serious, so I think at least minimal safeguards are necessary. But if that goes in a direction that kills innovation, I think we'll ultimately just lose competitiveness.
In the end, I think what matters isn't blocking the technology itself, but clarifying where responsibility lies. The problem, though, is that that's the hardest part.
5 answers
Agreed — what's most frustrating is that the people doing the regulating have such a low level of technical understanding. Many of them regulate deepfakes without even knowing how generative models work.
Well, the whole “regulation kills innovation” argument is getting a bit tired now. It’s a carbon copy of Big Tech lobbying rhetoric. And when an accident blows up because there were no safeguards, who’s going to take responsibility?
Clarifying where responsibility lies is definitely the answer, but we first need to point out why that's so difficult. Where to cut off liability—model developer, deployer, or user—varies case by case, so the law can't draw a clean line. That's why countries end up going with vague stuff like 'risk-tiered regulation' lol
This is exactly right lol. Regulation is about finding who's responsible, not sealing off the technology.