one year on
Seven AI Companies Sign Voluntary White House Commitments on Safety and Security
OpenAI, Anthropic, Google, Meta, Microsoft, Amazon and Inflection agree to red-teaming, watermarking and transparency measures in a pre-regulation handshake with the Biden administration.
Seven leading AI companies have announced voluntary commitments to manage the risks of artificial intelligence in a White House ceremony with President Biden today. The companies — Amazon, Anthropic, Google, Inflection, Meta, Microsoft, and OpenAI — agreed to internal and external security testing of AI systems before release, including red-teaming by independent experts. They also committed to sharing information across industry, government, civil society, and academia on managing AI risks, investing in cybersecurity to protect model weights, and developing watermarking mechanisms for AI-generated content.
The commitments are non-binding and come as the Biden-Harris administration works on an executive order and bipartisan legislation to govern AI. The White House described the measures as a way to ‘ensure that innovation doesn’t come at the expense of Americans’ rights and safety.’ The companies will also publicly report their systems’ capabilities, limitations, and appropriate and inappropriate use, and prioritize research on societal risks like bias and discrimination.
The move is part of a broader push by the administration to get ahead of AI regulation, having been caught flat-footed by social media’s disruptions. The agreement is non-binding and there is no enforcement mechanism — the attending leaders include Microsoft President Brad Smith, Google President Kent Walker, Anthropic CEO Dario Amodei, Inflection CEO Mustafa Suleyman, Meta President Nick Clegg, OpenAI President Greg Brockman, and Amazon Web Services CEO Adam Selipsky. No billionaires or women were present. The White House has already consulted on the voluntary commitments with allies including Australia, Canada, the UK, and Japan.
The record
The administration says the commitments underscore 'safety, security, and trust' and mark a 'critical step toward developing responsible AI.'
One year later — open only if you can handle spoilers
The voluntary commitments laid the groundwork for President Biden’s October 2023 Executive Order on AI, which imposed some of the same requirements on federal agencies. Within a year, several signatories had faced criticism over the scope of their red-teaming and watermarking efforts, and the pledge remained primarily a public-relations milestone rather than a binding regulatory framework.