In a bold statement this week, Solana co-founder Anatoly Yakovenko challenged the recent calls for a slowdown in artificial intelligence (AI) development, suggesting that the motivations behind this push are more rooted in profitability than in safety. His comments come in the wake of a public essay from Anthropic CEO Dario Amodei, who advocated for a more cautious approach to AI advancements.
Yakovenko’s succinct rebuttal on social media, which simply read, “Profitability at $1 trillion mcap,” raises eyebrows and questions about the underlying incentives driving the AI industry’s leading figures. While his post did not specify any particular companies or financial details, it implies a critique of the largest players in the field, suggesting their push for regulation is less about public safety and more about solidifying their market positions as they reach significant valuations.
Following Yakovenko’s remarks, he humorously indicated that he had instructed his Codex to use fewer tokens, subtly mocking the notion of artificially pausing AI progress. This stance aligns with the sentiment that the AI race should not be hindered by regulatory frameworks that could potentially stifle innovation.
Understanding Amodei’s Proposal
In his essay titled “We Must Pace the Frontier,” Amodei proposed a structured approach to slowing AI developments. He emphasized that the objective is not to halt research but to moderate the pace of capability improvements. His plan outlines three key stages, starting with allowing third-party evaluators continuous access to AI systems—similar to internal safety teams—highlighting organizations like METR, which focuses on model evaluation and threat research.
Amodei further suggested that AI companies in democratic nations should collaborate on shared safety requirements, acknowledging potential antitrust issues that may arise, and hinted at the need for limited government oversight. The final stage of his proposal involves international cooperation, potentially including agreements with nations like China, to establish controls on AI technologies, including those related to automated model enhancement.
OpenAI’s CEO Sam Altman quickly endorsed Amodei’s ideas, agreeing on the necessity of pacing AI advancements and emphasizing the importance of independent evaluators within AI companies. Elon Musk also expressed his support for Amodei’s views, although he refrained from detailing which aspects he particularly agreed with.
Pushed Back by Industry Voices
David Sacks, a former White House AI adviser, has voiced strong opposition to the idea of widespread regulatory frameworks. He believes that leading firms like Anthropic and OpenAI can self-regulate without dragging smaller entities into a restrictive environment. Sacks labeled these two companies as a “duopoly on frontier intelligence” and cautioned that top-down regulations could hinder smaller startups and open-source projects that are gaining momentum.
He further argued that achieving a global agreement on AI regulations is improbable, especially with the likelihood that countries like China would not adhere to such agreements. Sacks warned that imposing restrictions on American AI labs could result in a substantial technological disadvantage.
Despite the heated discussions, the markets appeared unfazed by the AI slowdown conversations, with analysts noting that a deceleration in frontier model development does not necessarily correlate with decreased investment in AI infrastructure as a whole.
As of now, Anthropic has committed to implementing outside evaluations but has yet to announce a specific timeline or evaluator for the process.
