I don't think they can convince me they have actually reversed course on this. Its invisible so we wouldn't know if they kept on doing it secretly. It required building out technical capability which is unlikely to remain forever unused while conveniently available to them.
They relied on trust that they were providing the service they were being paid for. That trust was blown, and an "oops, lets undo that" does not regain trust. It would be prudent to assume the invisible guardraild are possibly in play for all future Clause use, Fable or otherwise.
Yes they already had an accident where the model magically downgrades itself, very likely that it just produces less good output rather than just stops working isn’t it… my guess is they were testing these features, accidentally or not, and wrote up something to justify what people were seeing. I find it absolutely disgraceful I can’t trust it to learn ML any more without there being a chance it’s messing me around. This whole saga represents a huge loss of trust for me in Anthropic.
They relied on trust that they were providing the service they were being paid for. That trust was blown, and an "oops, lets undo that" does not regain trust. It would be prudent to assume the invisible guardraild are possibly in play for all future Clause use, Fable or otherwise.