Update (September 18, 2023): IBM, Adobe, Salesforce, Nvidia, Palantir, Stability AI, Cohere, and Scale AI joined the voluntary commitments pledge on September 12, bringing the total number of companies to fifteen. Original Story (July 25, 2023): Leading artificial intelligence (AI) companies including OpenAI, Google, Meta, Amazon, Microsoft Anthropic, and Inflection voluntarily committed to adopting the following measures in the development of AI technology, the US White House announced on July 21: Watermarking AI-generated content: The companies will develop technical mechanisms like watermarking to ensure that users are aware when content is AI-generated. This will enable "creativity with AI to flourish but reduces the dangers of fraud and deception," the press release stated. Publicly report on capabilities, limitations, etc: The companies will publicly report their AI systems' capabilities, limitations, and areas of appropriate and inappropriate use, covering both security risks and societal risks (effects on fairness and bias). Security testing of AI systems before release: The companies will carry out internal and external security testing of their AI system by independent experts before releasing it to the public. This will guard against risks like biosecurity and cybersecurity, the press release reasoned More research on societal risks that AI poses: The companies have committed to prioritizing research on the societal risks that AI systems can pose, "including on avoiding harmful bias and discrimination, and protecting privacy." Information sharing across the industry and with other stakeholders: The companies will share information on managing AI risks with others in the industry as well as with governments, civil…
