3 ms·
It's a feature not a bug. Maximizing banality is being cast as safety to avoid discussing the harder problems long term. I agree that any new technology comes w
by 65a 3y ago
It's a feature not a bug. Maximizing banality is being cast as safety to avoid discussing the harder problems long term. I agree that any new technology comes with risks, including serious ones, and humans should get better at managing those risks, but it's a long jump from that argument to "only we should have this new advanced technology, for your own safety", where safety is ill-defined and likely misses the very real safety risks of "only them having this new advanced technology"
- imjonse 3y agoI never argued for restrictions (I don't agree with the article authors) and it's a real risk if powerful AI is only controlled by already powerful entities. My comment was about the dangers of equating AI safety with aligning chatbots and censoring diffusion models.