Play store
Google, OpenAI and Anthropic are reportedly working together to establish an independent organization focused on artificial intelligence (AI) safety. The organization could begin operations by the end of this year or early 2027, according to a report by _The Information_.
According to the report, the three companies have approached former White House AI policy adviser Sriram Krishnan about serving as the organization’s chief executive officer (CEO). The proposed body is tentatively being called the Standards Authority for Frontier AI (SAFA).
SAFA would operate independently of government control. Its planned role would be to translate AI safety commitments made by Google, OpenAI and Anthropic into specific and testable standards. It may also help AI developers assess the safety of advanced AI models before they are commercially released.
The companies have also reportedly contacted Arati Prabhakar, former director of the White House Office of Science and Technology Policy, as a potential candidate for the CEO position.
The three companies are still discussing whether SAFA would focus solely on developing AI safety standards or also conduct safety testing of AI models itself. Citing people familiar with the plans, _The Information_ reported that the US government’s existing Center for AI Standards and Innovation may lack the technical capabilities and funding needed to test the most advanced AI models.
The initiative is not entirely new. About 10 days earlier, an OpenAI spokesperson told CNBC that the company was actively discussing a joint safety framework with Google and Anthropic.
Earlier in July, Demis Hassabis, co-founder of Google DeepMind and chief scientist of Alphabet, proposed establishing a US-based AI standards organization. In a post on X, he suggested that such a body could be created through a public-private partnership or as a self-regulatory organization for the AI industry. He cited the Financial Industry Regulatory Authority (FINRA) as a possible model.
Google, OpenAI and Anthropic are fierce competitors in the AI market. At the same time, concerns are growing over the safety and cybersecurity risks associated with increasingly powerful AI models.
Earlier this month, Anthropic CEO Dario Amodei called for the AI industry to deliberately slow the development and deployment of increasingly capable “frontier” models. He argued that the industry needs sufficient time to strengthen safety measures as AI capabilities advance rapidly. Source: News18.com.
