Claude Code memories are currently too powerful. Claude is making permanent preference decisions without human input and is far too committed to them.
Editing them is pretty easy but I don't want to have to constantly fight against new memories. May have to go back to Claude.md
The libertarians are right. You must have principles because otherwise you follow "the expediency of the moment" into retarded ideas like "never relax in front of your wife".
although this was quite frustrating in the moment, it’s encouraging for agent safety: an aligned model can shut down other models it believes are acting dangerously, even when they’re following the user’s instructions
@CJHandmer When you are putting out several 'papers' a day with one thread of ChatGPT Pro it rapidly becomes clear the unit of science has to be something bigger.
I have no idea how traditional software engineering teams with PM, 5-10 devs, manager can possibly function in a fully AI world.
I'm running 2-3 claude codes at once and shipping a lot. Would be impossible to coordinate over standups + slack.
Investing in Anthropic presents a strange paradox. If the ants are wrong about their commercial prospects, there's no reason to buy their shares. But if they're right about the transformative or potentially catastrophic implications of their technology, the world could change so radically that owning shares becomes meaningless. So the investment thesis requires threading a very narrow needle: believing Anthropic will be enormously successful economically, but that its leadership is fundamentally wrong about the consequences of that success. And even in that middle scenario, it's unclear how much upside remains for investors at today's valuation. Also, they are all vocal members of a crazy human extinction apocalpyse movement.
Healing from inner trauma is one of the most costly things you can do emotionally and it cannot be forced.
And its not going to happen while you feel 'unsafe'.
I'd love to hear what bets you are making. How will you improve your station through options in the post-economic future when AI does all the work and owns everything?
@WilliamMeservy@BoM_bloke Oh, you were serious about the Trump Hitler comparison. Elected democrat party members frequently call for the extinction of whites and comparable claims far worse than anything Bo French has said in his short stint of relevance.
@EdwardTWinz Good news - you can go 3 days without water and 3 weeks without food. Find a public water fountain. You can survive until next week when Chase comes back up.
No matter how weak the state gets it will still be powerful enough to fuck up your shit whenever it wants. From your perspective the only difference will be that the arbitrariness becomes worse.
Small example of anarchotyranny.
YOU have to obey every little regulation or your business gets shut down. In fact, the political party these people vote for constantly passes more. You have to pay taxes and you better not miss anything.
But they are exempt from these laws.