Tech companies make AI safety and transparency pledges at White House

Receive free Artificial intelligence updates

We’ll send you a myFT Daily Digest email rounding up the latest Artificial intelligence news every morning.

US technology companies including Google and OpenAI publicly committed to promoting safety and transparency in the development of artificial intelligence at the White House on Friday.

Executives from Amazon, Anthropic, Google, Inflection AI, Meta, Microsoft and OpenAI made “voluntary” commitments “to help move toward safe, secure and transparent development of AI technology”, the White House said.

The commitments — which include agreeing to internal and external security testing of AI systems before they are released to the public — come less than three months after the Biden administration convened tech executives at the White House for what was described as a “frank discussion” about safety concerns relating to AI.

Several executives appeared at the White House alongside president Joe Biden to tout their new public undertakings, including Microsoft president Brad Smith, Inflection AI chief executive Mustafa Suleyman and Nick Clegg, the president of Facebook and Instagram parent Meta.

“Artificial intelligence [poses] . . . both risk to our society and our economy and our national security but also incredible opportunities,” Biden said.

Clegg called the voluntary commitments an “important first step in ensuring responsible guardrails are established for AI”. Smith said the White House had “create[d] a foundation to help ensure the promise of AI stayed ahead of its risks”.

Anna Makanju, vice-president of global affairs at OpenAI, said the commitments “contribute specific and concrete practices” to the “ongoing discussion” around regulating AI. OpenAI chief executive Sam Altman testified before Congress in May and called for lawmakers to boost regulation of AI, warning: “I think if this technology goes wrong, it can go quite wrong.”

In addition to committing to more security testing, the companies also pledged to share more information across the industry and government about how they are mitigating risks. They also said they would invest more money on cyber security safeguards, and promised to make it easier for third parties to discover and report vulnerabilities in their systems, the White House said.

The Biden administration and lawmakers on Capitol Hill have scrambled to formulate a coherent policy response to rapid advances in AI technologies that have emerged in recent months.

The White House called the voluntary commitments a “critical step toward developing responsible AI” but noted that the administration was still preparing an executive order, and urging Congress to pass legislation, to further regulate the development of AI.

A White House official said the voluntary commitments were “pushing the envelope on what companies are doing and raising the standards for safety, security and trust of AI”. However, that did not “change the need” for bipartisan legislation and an executive order from the White House.

“This is a high priority for the president and the team here,” the official added.

“In the weeks ahead I’m going to continue to take executive action to help America lead the way towards responsible innovation,” Biden said on Friday. “We’re going to work with both parties to develop appropriate legislation and regulation.”