OpenAI's GPT-4 and Anthropic's Claude 3 Sonnet can be tricked into generating unsafe content - quite easily.
Posts tagged as “Responsible AI”
The NTIA is advising the White House to let advanced AI models openly share their key components, but also monitor them - and put on restrictions if the risks outweigh the benefits.
OpenAI's former Chief Scientist Ilya Sutskever has started a new company called Safe Superintelligence Inc. that puts safety ahead of AI development.
Eleven OpenAI employees and two from Google DeepMind signed an open letter saying leading AI companies cannot be trusted with keeping society safe from AI's harms.