Tech Leaders Seek Antitrust Exemptions to Coordinate on AI Safety

Leaders at leading artificial intelligence companies are pushing for a special antitrust exemption that would permit them to coordinate on safety standards and development pace. This proposal has sparked a sharp debate in Washington and Silicon Valley, with critics arguing that such an exception would effectively allow the industry to form a cartel, limiting the ability of smaller rivals to compete.

The push comes as several major labs face internal dissent over safety protocols, with some researchers leaving their posts in protest. Meanwhile, the U.S. government’s approach to regulating these powerful technologies remains fragmented, leaving the industry to self-manage risks that experts say could have significant economic and societal consequences.

Coordinating safety creates competitive barriers

According to Jonathan Kanter, former chief of the U.S. Department of Justice Antitrust Division, the request for an exemption is fundamentally about market structure. In an interview with GN technics/ai (en-US), he explained that allowing competitors to meet and agree on safety measures removes the pressure to innovate on cost and efficiency. Instead of racing to build better products, firms would focus on maintaining high barriers to entry.

Kanter noted that this coordination could shield established players from cheaper alternatives, particularly open-weight models developed in China. By setting industry-wide standards, incumbents can ensure that only they have the resources to comply, effectively locking out new entrants who might otherwise disrupt the market with lower-priced solutions.

Regulatory vacuum complicates oversight

The current regulatory environment is described as a landscape without rules, similar to a road system with no traffic lights or speed limits. With the federal government currently taking a hands-off approach, companies are left to define their own compliance frameworks. This lack of clear external oversight makes it easier for large firms to interpret safety requirements in ways that favor their existing business models.

Kanter argued that without independent antitrust scrutiny, the industry’s self-regulation may not address public safety concerns adequately. The trade-off is clear: while coordination might slow down reckless deployment, it also risks creating a static market where innovation is driven by corporate strategy rather than consumer benefit.

China competition shapes policy stance

The debate is further complicated by geopolitical tensions. U.S. policymakers are concerned that overly restrictive regulations could cede the advantage to Chinese competitors, who are aggressively deploying open-source AI tools. Conversely, too little regulation could allow a few U.S. giants to dominate the market, reducing the diversity of approaches and potential breakthroughs.

Kanter suggested that the path forward requires a balanced approach that maintains competitive pressure while ensuring safety. He emphasized that antitrust law is designed to preserve the dynamic process of competition, which is the primary engine for technological progress. Removing that pressure, even for noble causes like safety, carries the risk of long-term stagnation.