Artificial Intelligence Companies Committed to Pre-launch Security Measures

The White House said Friday that it has secured voluntary commitments from seven US companies aimed at securing their artificial intelligence products before they launch. Some commitments require third-party oversight of how commercial AI systems operate, with no clarity on who will audit the technology or hold companies accountable.

The skyrocketing commercial investment in generative AI tools that can write compelling human text and create new images and other media has sparked public admiration and concern about their ability to deceive people and spread misinformation, among other dangers.

Commitments to Address Major Risks

The White House said in a statement that the four tech giants, along with (OpenAI), app developer (ChatGBT), and startups (Anthropic) and (Inflection AI), have committed to security testing “partially conducted by independent experts” to protect against major risks such as biosecurity and cybersecurity.

The companies also committed to using vulnerability reporting methods on their systems and using digital watermarks to distinguish between real and AI-generated images.

The White House said companies will publicly report flaws and risks in their technology, including the impact on fairness and bias.

Immediate Measures and Long-Term Regulation Efforts

Voluntary commitments are intended as an immediate way to eliminate risks before a long-term effort to get Congress to pass laws to regulate this technology.

Some advocates of AI regulation have said that Biden’s move is only the beginning, but more needs to be done to hold companies and their products to account.

In a statement, James Steyer, founder and CEO of the non-profit organization Common Sense Media, said: “History shows that many technology companies do not actually live up to their voluntary commitment to act responsibly and abide by strict rules.”

Senate Majority Leader Chuck Schumer, a New York Democrat, has signaled that he will introduce legislation to regulate artificial intelligence. He held a series of briefings with government officials to brief senators on a bipartisan issue.

A number of tech executives have called for AI regulation, and some of them traveled to the White House in May to speak with Biden, Vice President Kamala Harris and other officials.