Post

"Anthropic’s commitments under the EU AI Act’s Code of Practice" A reminder that these things are not really that complicated. We are not the serfs of tech billionaires. We can regulate to get the world we want to live in. And it is easier to regulate if you operate at scale, as the EU does.

Jo Michell@jomichell.bsky.social

This is interesting: Claude is going to add 'watermarks' to its output. support.claude.com/en/articles/...

I know people will yell at me but: "Regulate the shit out of it and fine companies a bazillion dollars when they're negligent or willfully harmful" seems like a more sustainable position in the long term than "reject all AI on principle." I just don't think the latter argument is winnable.

14 reposts170 likes

I agree sadly the continued rise of AI is inevitable (sorry to sound all Sarah Conners on this) so regulation is the best way to adapt.

Pragmatism over absolute rejection: outright tech bans never hold, but regulation only works if penalties actually hurt. Fining tech giants pocket change won't stop negligence; enforcement needs teeth that disrupt business models to make compliance cheaper than non-compliance.

I just don't see either actually happening. People sometimes cite the ozone layer as proof we can solve problems with regs and enforcement and what I think is a) that was awhile ago, and b) industry's takeaway was "never again."

What happened w/ the ozone layer was a combo of: This is really bad, & it's really bad right now & This isn't that hard to fix, & the fix won't require massive changes to infrastructure

the tech is being developed by techbros and some of the worst and greediest corporations on the planet that absolutely don’t give a shit about you or anyone else. the option to develop this ethically never left the table they just picked the worst choice because the hold everyone in contempt.