Oh my god have Anthropic products been absolutely saying everything is load bearing for the last week or so. Literally ever other paragraph has “such and such is load-bearing”.
> You're proving my point. The thumb was put on the scale, he public was bombarded 24/7 with self-serving false dichotomies and viola, you've just manufactured mass public support for insane bullshit.
See also Covid-19. Same shit only waaaaaaay more batshit insane and waaaaaaaay more crazy 24-7 fear mongering.
Tool discovery is only one very small part of the “context problem”. The bigger problem is the outputs. You can’t compost them. Ever watch Claude code run some crazy shit through python and jq to take some input, transform it in some crazy way and output exactly what it needs back into its context? You simply can’t do that with mcp. It’s basically forced to accept the exact shape of the mcp output into its context and then take that intermediate output and dump it right back into another tool. That is incredibly wasteful!
If your lucky the mcp might expose a way to ship its output into a text file so at least the agent can have a go at it with CLI tools.
This is actual reason. So any investors reading our system card.... write us another check and watch the $$$$$$$$ roll in. It's so dangerous we can't even release it!
"for example, consider using models to write email -- is it a misalignment problem if the model is just too good at writing marketing emails?? or too good at getting people to pay a spammy company?"
But who gets to be the judge of that kind of "misalignment"? giant tech companies?
Funny enough, today it seems to have stopped…
reply