Anthropic's Horrible New Restrictions
This title could be clearer and more informative.Try out Clickbait Shieldfor free (5 uses left this month).
Anthropic's new 'Fable' mythosclass model is drawing criticism for its restrictive content policies. Despite strong benchmark performance overall, the model reportedly refused all 200 tasks in certain benchmarks like 'program bench,' falling back to Opus 4.8 in other cases. The author argues these restrictions are unacceptable and set a troubling precedent for AI model content policies.
•1m watch time
1.6K Impressions