Biden says AI leaders committing to building ‘safe, secure and trustworthy’ tech

President Joe Biden on Friday said the burgeoning artificial intelligence sector can provide “incredible opportunities” for innovation as he announced voluntary commitments by leading AI companies to manage risks posed by the fast-developing technology, which has already spawned a growing number of popular applications such as OpenAI’s ChatGPT chatbot.

Speaking from the Roosevelt Room of the White House where he was joined by chief executives from seven American AI leaders, Mr Biden said the companies — Amazon, Anthropic, Google, Inflection, Meta, Microsoft, and OpenAI — have all agreed to what he described as “voluntary commitments for responsible innovation”.

The president said the commitments would be implemented “immediately,” and would be focused on “three fundamental principles: Safety, security and trust”.

According to a fact sheet distributed by the White House, each of the companies has agreed to commission internal and external security tests of new AI systems before making them available to the public. The tests are meant to protect against “significant sources of AI risks,” including biosecurity and cybersecurity risks as well as “broader societal effects”.

Each of the companies has also agreed to invest in security safeguards to protect “model weights” that are proprietary or unreleased, as well as to develop “robust technical mechanisms” to “reduce the dangers of fraud and deception” by permitting AI-generated content to be easily identified through technical means such as watermarking.

Mr Biden also said the companies’ investment in trustworthy AI technology will include empowering users to detect altered or fake content, as well as “rooting out bias and discrimination, strengthening privacy protections and shielding children from harm”.

“These commitments are real and they're concrete. They're going to help fulfill industry fulfill this fundamental obligation to Americans to develop safe, secure and trustworthy technologies that benefit society and uphold our values,” he said.

The president said the rise of AI has been an “astounding revelation” for him, and predicted that Americans would see more technological change in the next decade than in the previous five, and praised the executives as having “critical” roles in “shepherding that innovation with responsibility and safety ... to earn the trust of Americans”.

“Social media has shown us the harm that powerful technology can do without the right safeguards in place ... we must be clear-eyed and vigilant about the threats emerging from emerging technologies that can pose don't have to but can pose to our democracy and our values. Americans are seeing how advanced artificial intelligence and the pace of innovation have the power to disrupt jobs and industries,” he said.

Mr Biden added that the “promising step” of the companies’ voluntary commitments does not rule out the need to do “a lot more work” to manage risks of AI with “new laws, regulations and oversight,” and promised that his administration would take further executive actions to “help America lead the way towards responsible innovation” while working with both parties in Congress to develop “appropriate legislation and regulation”.

“This is a serious responsibility,” he said. “We have to get it right”.