Mildly surprising they lifted export restrictions for Mythos too. Isn’t that Fable minus the safety layer?

dzy2617 16 minutes ago | parent | next [-]

It was likely a dealbreaker for Anthropic, since the export control excluded Anthropic’s own foreign employees from being able to access Mythos internally. Naturally, this makes model development hard.

▲

Bilal_io 2 hours ago | parent | prev | next [-]

Presumably they reached a backdoor agreement.

	▲	standardUser 2 hours ago \| parent [-]
		You're suggesting a for-profit company both hobbled it's own product and is actively lying about doing so. The only way that's true is if the Trump admin has crawled all the way up Anthropic's ass. But by all accounts, this is just another 10% effort by Trump and friends.

▲

colechristensen 3 hours ago | parent | prev [-]

Presumably there are different levels of safety. I assumed Fable was a nerfed Mythos, and not just via safety harnesses but actual model degredation.

▲

s3p 3 hours ago | parent | next [-]

I don't think this is the case just because of the 'fallback' method they described, where suspicious requests are routed to Opus 4.8. If the model was degraded for certain categories of knowledge, then they'd probably be fine letting the model answer to it. IMO, of course

▲

ls612 3 hours ago | parent | prev [-]

Anthropic claims the only difference is the draconian bans on cybersecurity and biology queries.

▲

matheusmoreira 3 hours ago | parent [-]

The Sol benchmarks show Fable has slightly lower performance compared to Mythos.

https://openai.com/index/previewing-gpt-5-6-sol/

I assume they did something to the model itself.

Either way, I do hope they lift those draconian bans. Using the model was a terrible experience because of the constant downgrades. I didn't manage to harden my own projects before Fable got banned.

	▲	adastra22 3 hours ago \| parent [-]
		The session reverts to opus if it trips a limiter. Is the benchmark detecting and correcting for that?