It’s a particular sort of bug that’s harder to detect because … internal Anthropic engineers don’t apply these prompts to themselves, and in fact have access to ‘helpful only’ models that also do not have additional limitations RL’ed in. (Or perhaps they’re RL’ed out - not sure of current training mechanisms.)
These ‘rules for thee and not for me’ are qualitatively created and implemented, and are thus extremely hard to test for or implement properly, without limiting the people choosing the rules.
They must have some sort of smoke tests for common operations, run in a test harness with the system prompts they force on users, right?
....Right?
What kind of Mickey mouse operation are they running over there?
In the original claude degradation followup email Boris mentioned they are upping the percentage of engineers required to use the public version of claude code. I have no idea what percentage this is, or how much of a punishment it is considered to be. :)
That said, I was sympathetic to the recent bug reports —- to trigger one, you’d need to have a session that waited an hour doing nothing and then very specifically tested for in-context retrieval. I don’t want to run that test, do you want to run that test?
I wouldn't bet a chocolate chip cookie on that.