Faced with community accusations of covert sabotage, Anthropic apologizes and cancels the secret downgrading restrictions on Claude.

Faced with community accusations of covert sabotage, Anthropic apologizes and cancels the secret downgrading restrictions on Claude. According to Beating Monitoring, Anthropic announced an adjustment to the development safety strategy of its new model Claude Fable 5, removing the restriction on silent performance downgrades.
The silent downgrade mechanism has been criticized by the community as "covert sabotage," leading to strong backlash from the AI research community.
Next, three things need to be monitored: whether related funds continue to flow in, whether on-chain transaction volume and holdings continue to grow, and whether the project team or regulators provide new confirmation information.
A single news flash can only indicate that sentiment has been ignited; subsequent data will determine whether it can develop into a trend.
Risks should also be considered: According to Anthropic's terms of service, users are not allowed to use Claude to train competing models.
Anthropic plans to directly reduce the performance of Claude Fable 5 for accounts suspected of training competing models, without notifying the users.
#链上数据 #AI #监管 #Blockchain #CryptoMarket