OpenAI, Google, others pledge to watermark AI content for safety, White House says

Reuters reports that AI companies, including OpenAI, Alphabet, Meta Platforms, Anthropic, Inflection, Amazon.com, and Microsoft, have made voluntary commitments to the White House to implement measures such as watermarking AI-generated content to help make the technology safer. President Joe Biden, who hosted executives from these companies, emphasized the need for vigilance about the threats from emerging technologies to U.S. democracy. The companies pledged to thoroughly test systems before releasing them, share information about how to reduce risks, and invest in cybersecurity.

The U.S. lags the EU in tackling artificial intelligence regulation. In June, EU lawmakers agreed to a set of draft rules where systems like ChatGPT would have to disclose AI-generated content, help distinguish so-called deep-fake images from real ones, and ensure safeguards against illegal content.

As part of the effort, the seven companies committed to developing a system to “watermark” all forms of content, from text, images, audios, to videos generated by AI so that users will know when the technology has been used. This watermark, embedded in the content in a technical manner, presumably will make it easier for users to spot deep-fake images or audios.