In a statement made on Tuesday, Dario Amodei, the CEO of Anthropic, an AI startup, described the AI Action Summit held in Paris this week as a “missed opportunity.” He emphasized the need for both the AI industry and government entities to “accelerate their efforts with greater precision.”

“We were grateful for the chance to participate in the AI Action Summit in Paris, and we recognize the French government’s initiative in assembling AI companies, researchers, and policy experts globally,” Amodei noted. “However, a stronger sense of urgency is necessary on various issues due to the rapid evolution of technology.”

Amodei’s critique of the AI Action Summit resonates with opinions expressed by multiple academics earlier this week. One expert remarked to Transformer that the commitments emerging from the conference, which were not endorsed by the U.S. and U.K., amounted to “nothing but empty rhetoric.”

In his remarks during the summit, U.S. Vice President JD Vance took a contrasting approach, criticizing what he termed “overreaching” regulations on AI proposed by European nations. He also expressed concerns regarding content moderation, questioning the “sustainable” and “inclusive” language in the conference’s agreements, which he viewed as forms of “authoritarian censorship.”

Amodei cautioned that the rapid advancement of AI brings significant risks, and neglecting to implement adequate regulations could lead to serious repercussions.

“The evolution of AI systems should be seen as akin to a new nation populated by highly intelligent individuals entering the global arena,” Amodei explained. “The emergence of advanced AI poses considerable threats to global security, including potential misuse by non-state actors… It is crucial that democratic societies take the lead in this field to prevent authoritarian regimes from achieving military dominance via AI.”

He urged governments to mobilize resources to monitor AI usage and formulate policies to ensure that the economic benefits of powerful AI technologies are shared equitably. Additionally, he advocated for enhanced transparency regarding AI safety and security, along with strategies to evaluate AI-related risks.

Amodei’s evaluation of the Paris AI Summit’s outcomes contrasts with that of OpenAI, which stated in a recent statement that the event marked “another significant step towards the responsible and beneficial advancement of AI for all.”

Anthropic has typically embraced the idea of AI regulation. Notably, Amodei has previously issued similar warnings in the past, highlighting that unchecked AI development could have far-reaching negative implications for economy, society, and security. Anthropic was among the few AI organizations that supported California’s SB 1047, a comprehensive and highly debated AI regulation bill. In contrast, OpenAI opposed this bill, which was ultimately vetoed by Governor Gavin Newsom last fall.

However, it’s important to note that Anthropic’s interests may not be entirely altruistic. Similar to OpenAI’s Sam Altman, who discussed the distribution of AI benefits in a recent essay, Amodei does not provide explicit suggestions for ensuring that the advantages of advanced AI, if it materializes soon, are distributed fairly and widely.

Explore our comprehensive coverage of the Artificial Intelligence Action Summit in Paris.