What Can't Fable Do?
I'm curious, has anyone encountered scenarios where Fable wasn't able to solve a problem, fix a bug, complete some task, etc. outside of the (questionable) restrictions in place? Every conversation I read about Fable devolves into either two conclusions: (1) Fable is insane and was able to do X in one or two prompts or fix issues that other models and humans had been grinding on for a while or (2) Fable sucks because my prompt triggered the guardrails. But I haven't seen much of "Fable wasn't able to do X" and in my work so far I haven't seen it really fail or mess up much at all.
I asked Fable if i should go to Small Cheval in Rosewood before my flight. It told me that I should go to the one in terminal 3 instead, its right by the k gates at the food court. I got there and.. there was no Small Cheval.. seems like Anthropic still hasn't figured out how to prevent halucinations...
"If you've never seen Fable fail, it might not be because it lacks boundaries, but rather because you haven't put it through a sufficiently real, complex, and verifiable production environment.
claude-fable-5 failed for me to follow instructions from CLAUDE.md to create a TODO list https://github.com/marcindulak/claude-fails-to-follow-claude...