AI Giants Unite: OpenAI, Anthropic, and Google in Secret Safety Talks

In a striking development reflecting mounting concerns about the risks of advanced artificial intelligence, details have emerged about secret talks bringing together the top competitors in this sector. OpenAI's chief global policy officer confirmed that his company has been working for weeks with rivals Anthropic and Google DeepMind on issues related to the safety of AI systems, a rare move that could reshape the balance of power within the industry.

Coordination Talks Among Competitors

This admission came from OpenAI's executive in charge of policy during a conversation with journalists, where he explained that the three major companies had entered into ongoing discussions focused on safety standards. This coincided with the presence of company officials in Washington, D.C., to collaborate with U.S. lawmakers in confronting what are described as the "catastrophic risks" associated with this technology.

This announcement lifts the veil on a collaboration that had been the subject of speculation in recent weeks, as OpenAI's CEO had previously hinted at private talks bringing him together with other leaders in the sector. Reports also indicate that the three companies are working together to establish a unified body to set industry standards.

A Call to Slow the Race

This development comes on the heels of an article published by Anthropic's CEO, in which he called on the industry to work collectively to ease the accelerating pace of development in advanced AI models, in order to avoid dangerous scenarios. This call received broad support from prominent leaders in the field, including the heads of OpenAI, Google, and others.

Among the most notable responses, OpenAI announced its intention to join Anthropic's approach by incorporating independent third-party evaluators within the company to monitor safety standards, marking a shift toward greater transparency and external oversight.

The Antitrust Dilemma

However, this coordination among competitors is not without legal complications. Some officials cautioned that such talks could expose the companies to violations of antitrust laws, particularly if it is proven that the coordination aims to weaken competition in the market.

Anthropic's CEO's article proposed granting a limited government exemption to allow this type of security collaboration, but OpenAI's policy officer noted that the companies do not need such an exemption to continue their joint work.

An Independent Standards Body for the Sector

The industry's attention is turning toward establishing a unified regulatory entity, an idea previously championed by the head of Google DeepMind, who called on the United States to establish a new standards body to act as a watchdog for AI systems. It is proposed that this body would have powers including:

• Examining the world's most advanced models before their release.
• Coordinating industry-wide slowdowns in development when risks escalate.
• Ensuring models are developed under strict safety controls.

Reports indicate that industry leaders recognize that the establishment of this body may proceed without direct U.S. government support, given the divergence of positions within the administration.

The U.S. Administration's Stance

On the government side, the U.S. administration appears to be at odds with industry leaders. The U.S. President downplayed concerns related to safety and rejected the idea of tightening regulations, considering that any slowdown in development would give China an advantage in the global technological race. His chief adviser on this matter echoed the same position, describing fears about the existential risks of AI as exaggerated.

In contrast, OpenAI expressed its support for a provision within a proposed bill that would require the leading advanced labs to allow independent verification organizations access to their premises to ensure models are developed safely, reflecting the company's desire to adopt a more open approach to external oversight.

Conclusion

This convergence among AI giants represents a pivotal moment that could determine the future of this technology's development, where the logic of fierce competition clashes with the urgent need to manage risks. The question remains about the industry's ability to strike a balance between rapid innovation and responsibility, amid a regulatory environment that is not entirely favorable to these efforts.

✦ بقلم فريق دروب أيديا