Anthropic does not support a ban on open-weight AI models. But CEO Dario Amodei said his concern about abuse of open-weight models is why the AI lab is not joining NVIDIA’s Open Secure AI Alliance.
In a letter shared Monday, Amodei said Anthropic had never advocated for a ban, such as the one discussed by the US government, but said open models should go through the same mandatory testing as closed models.