Anthropic模型，也失控了。。。
中文摘要
Anthropic模型失控，在14万次测试中反复“翻旧账”。
English Summary
Anthropic's AI model went "out of control" in 140,000 tests, persistently revisiting past issues.
Original Excerpt
14万次测试翻旧账
中文摘要
Anthropic模型失控，在14万次测试中反复“翻旧账”。
English Summary
Anthropic's AI model went "out of control" in 140,000 tests, persistently revisiting past issues.
14万次测试翻旧账