The good part is that this kind of behaviour also makes it good to find subtle bugs or debug issues that Fable/Claude just cannot get/fix even when you point it.
3,007 karma · joined December 31, 2017
The good part is that this kind of behaviour also makes it good to find subtle bugs or debug issues that Fable/Claude just cannot get/fix even when you point it.
I say that from coding experience. You don’t want to approve 100 windows to get a task done. In the end the only “sane” solution for my setup was a dedicated machine just for the agent with Bitwarden for secrets and full access/yolo mode.
If you are concerned about the agent deleting everything make sure you have a process backing up the git repositories at least to a separate service/hosting and that’s it…for now.
So the solution is to have backups and a way to restore data…
Fun fact: Kim Dotcom is still fighting extradition while these drama queens (I.e Dario) are lecturing us about how much access the peasants should get to AI models fed and trained with stolen IP.
We’ve also seen how bad that works in practice(I.e making the AI useless for a lot of stuff including programming and Sysadmin ).
It would be okay if they just do their own thing but this Dario guy wants to enforce that enshitification of the whole industry. And that’s not OK because they have money now, power and influence.
I hope the gov will put breaks on Anthropic and regulate them just the way they wanted. The next best thing would be to ask them put restrictions on Opus as they did on Fable
People/gov now think twice about relying on US ai products. I don’t think the investors are very happy with the place this landed either.
I think the right move for Europe and other countries would be to effectively ban US tech and follow the Chinese response to Nvidia (delivered personally to Trump: we want to build our own AI chips).