News for hackers. Stuff that matters.
Re:Why Is Claude Turning into an a**Hole? by bombcar on Sunday June 14, 2026 @07:05PM (#48533954)
That’s the problem - it seems like a mind but it doesn’t operate like the ones we’re used to.

Even a dog will learn from recent stimuli, these things don’t. The prompt just modifies.

Re:Why Is Claude Turning into an a**Hole? by whstl on Sunday June 14, 2026 @07:21PM (#48534121)
I don't think that's a problem here at all.

The problem here is not doing tasks and outputting garbage output.

Re:Why Is Claude Turning into an a**Hole? by operatingthetan on Sunday June 14, 2026 @07:14PM (#48534034)
Yep, the only way these things can have "memory" is by shoving previous conversations into the context window.
Re:Why Is Claude Turning into an a**Hole? by coldtea on Monday June 15, 2026 @03:08AM (#48537618)
That's only because we hardcoded their weights in our implementation.

Aside from the cost and slowness, nothing about an LLM prevents feeding recent stimuli in and using it to update the models/retrain.

Re:Why Is Claude Turning into an a**Hole? by coldtea on Monday June 15, 2026 @03:08AM (#48537615)
That's only because we hardcoded their weights in our implementation.

Aside from the cost, nothing about an LLM prevents feeding recent stimuli in and using it to update the models/retrain.

One can even do it in a makeshift way without modifying the weights, just keeping a complete version of any prompt + vector search on disk memory of it.