Microsoft Google Alliance actively promotes AI Security Standards

Microsoft, Google, artificial intelligence start-ups OpenAI and Anthropic are stepping up efforts to develop artificial intelligence (AI) safety standards and today appointed a director to fill a global regulatory gap.

This summer, the four tech giants joined forces to form a new industry organization, the Frontier Model Forum, dedicated to developing AI models safely and responsibly. Today, they appointed Chris Meserole, director of artificial intelligence at the Brookings Institution, an American think tank, as executive director of the organization.

In addition, the Frontier Model Forum said it plans to invest $10 million in an AI security fund.

Meserole said: "We may still be some way away from real regulation. In the meantime, we want to make sure these systems (regulatory rules) are as safe as possible. "

Over the past few months, there has been growing concern that increasingly powerful artificial intelligence has the potential to replace human jobs, create and disseminate misinformation, or eventually surpass human intelligence.

Meserole said the forum would seek to "supplement" any official regulations. The EU AI bill is expected to be finalised early next year. Meanwhile, the UK will host the first Global AI Security Summit next week, where world leaders and key tech executives will be invited to discuss relevant cooperation.

Meserole also said the forum will initially focus on risks, including the ability of artificial intelligence to help design biological weapons and the ability to generate computer code that can be used to facilitate hacking of critical systems.