Search
Anthropic CEO Dario Amodei believes that governments should have the authority to block the deployment of new artificial intelligence models if they pose specific risks. He advocates for mandatory third-party testing of AI models to assess their risks across multiple domains. He compares the current stage of AI to automobiles, airplanes, or pharmaceuticals—technologies that are crucial to the economy but can cause significant casualties if improperly designed or used. Previously, Anthropic warned that its Mythos model could discover and exploit critical software vulnerabilities, so it was only made available to a small number of partners. This week, the company released a new version with the relevant cybersecurity attack capabilities removed.
-
C114 Communication Network -
Communication Home
