Top A.I. Executives Agree to White House Safeguards

The U.S. has not yet passed comprehensive federal legislation to regulate AI, so holding these companies to their promises will be difficult.

Greg Brockman sits wearing a blue shirt.
Greg Brockman, president of OpenAI, is reportedly at the White House today. Getty Images for SXSW

A group of seven companies leading the development of artificial intelligence have agreed to follow a list of safeguards recommended by the White House, the Biden administration announced in a release today (July 21). Companies include Meta (META), Google (GOOGL), Amazon (AMZN) and Microsoft (MSFT). The commitment is immediate, but it is voluntary and non-binding, so it isn’t legally enforceable.

As part of the agreement, the companies will participate in an independent review of their AI systems’ security before they release the technology publicly. Companies have also agreed to share information about managing the risks of AI across the industry, with government and researchers, according to the release. OpenAI, Anthropic and Inflection AI have also pledged to follow the White House’s safeguards.

The launch of OpenAI’s ChatGPT in November 2022 sparked rapid developments in the AI field, but some have grown increasingly wary that the technology is moving too quickly. In March, more than 1,000 experts called for a six-month pause on AI developments to study the risks and implement safeguards. Two months later, a group of Big Tech executives, AI scientists and researchers signed an open letter stating that AI advancements could lead to human extinction.

The White House’s action is part of a larger campaign to address the safety and security of AI technologies. In May, Vice President Kamala Harris and senior officials hosted CEOs of major AI companies—including Google’s Sundar Pichai, Microsoft’s Satya Nadella, OpenAI’s Sam Altman and Anthropic’s Dario Amodei—to discuss the safety of artificial intelligence. The Biden administration has also published a series of documents outlining principles for responsible AI innovation. The U.S. has not yet passed comprehensive federal legislation to regulate AI, so holding these companies to their promises will be difficult.

Who is involved?

Representatives from each company will meet at the White House today, according to NBC. They include Brad Smith, Kent Walker, Adam Selipsky, Nick Clegg, Greg Brockman, Dario Amodei and Mustafa Suleyman.

  • As Microsoft president, Brad Smith is in charge of making the case for the company’s products publicly. Smith’s background is in law, and he joined Microsoft three decades ago in the legal and corporate affairs department. He worked as general council for 13 years and advanced to the president role in 2015. He is also a member of Netflix’s board of directors.
  • Kent Walker is the president of global affairs at Google, where he manages the company’s communication with governments around the world. Walker joined Google in 2006 as vice president and general counsel. Similar to Smith, Walker’s background is in law. He formerly worked at the U.S. Department of Justice from 1990 to 1995.
  • Adam Selipsky became the chief executive of Amazon Web Services two years ago. He previously built the division’s marketing and sales strategy from its founding in 2006. In 2016, he left to become CEO of Tableau Software. He led the software company to acquisition from Salesforce in 2021 before returning to Amazon Web Services.
  • Meta’s president of global affairs, Nick Clegg, formerly worked in British politics. Since joining Meta in 2018, Clegg has become a prominent spokesperson for the metaverse and AI. He recently published an article making the case for Meta’s approach to open-sourced AI.
  • Greg Brockman co-founded OpenAI in 2015 alongside Sam Altman and Elon Musk, and he now works as president of the company. He previously dropped out of the Massachusetts Institute of Technology to join Stripe, the financial services company, where he worked as its first chief technology officer.
  • Anthropic is the AI startup behind the chatbot Claude, founded in 2021 by former members of OpenAI. CEO Dario Amodei was one of the founding members. He spent four years at OpenAI, working in positions that focused on research and safety. Amodei left the company due to “differences over the group’s direction” following its $1 billion investment from Microsoft, according to the Financial Times.
  • Inflection AI CEO Mustafa Suleyman co-founded the company last year after developing the chatbot Pi. Suleyman formerly founded DeepMind, another AI company that Google acquired for more than $500 million in 2014. Suleyman worked at Google as vice president of AI product management and AI policy for two years.

Top A.I. Executives Agree to White House Safeguards