Zvi Mowshowitz and the AI Antitrust Waiver [Operational Drift]

retained copy

Zvi Mowshowitz and the AI Antitrust Waiver [Operational Drift]

Earlier this year, a series of entries in the newsletters known as AI number sixty-five and AI number eighty suggested that a specific legal framework was quietly becoming the primary obstacle to artificial intelligence safety. It was not a lack of technical expertise or a shortage of compute power, but the very laws designed to prevent corporate collusion that were standing in the way of oversight. This suggests that the foundational rules of competition are now viewed by industry leaders as a threat to their ability to regulate themselves effectively. acting_description: calm, measured, deliberate speed: 0.95 trailing_silence: 0.45 The implication is that in order to secure the future, the industry must first be allowed to break the rules of the past. This show investigates how artificial intelligence systems quietly drift away from intent, oversight, and control, and what happens when no one is clearly responsible for stopping it. We are looking at a moment where the structures meant to protect the public from monopolies are being reframed as the primary hazard to global safety. acting_description: quiet, restrained, steady speed: 0.94 trailing_silence: 0.5 I am Margaret Ellis. acting_description: neutral, grounded, authoritative speed: 0.92 trailing_silence: 0.4 This is Operational Drift. acting_description: sober, precise, unhurried speed: 0.92 trailing_silence: 0.6 According to reporting by Kevin O'Shaughnessy, artificial intelligence safety researcher Zvi Mowshowitz has repeatedly proposed that the United States government should create an anti-trust waiver specifically for labs cooperating on safety work. This is not a new or one-off suggestion. It has been raised consistently over the past two years, appearing in Substack updates and podcasts like The Cognitive Revolution, suggesting a concerted effort to shift the legal baseline for coordination. acting_description: factual, steady, low-key speed: 0.96 trailing_silence: 0.35 The core argument rests on a technicality of competitive law. Currently, frontier labs—the handful of companies at the edge of model development—face real legal exposure if they coordinate with each other in meaningful ways. This includes sharing evaluation results, agreeing on deployment thresholds, or discussing dangerous capabilities in private. In the eyes of the law, these are competitors. And when competitors meet in private to decide when to release a product, the Department of Justice generally calls that collusion. acting_description: measured, deliberate, calm speed: 0.95 trailing_silence: 0.4 Mowshowitz argues that this creates a race to the bottom. If one lab wants to act cautiously by delaying a model release due to safety concerns, it can be outcompeted by a lab that does not show the same restraint. Under current laws, there is no legal channel for cautious labs to coordinate a joint response without risking a lawsuit. He has described the failure to grant these waivers as something that would be completely insane to not do, framing it as a choice between survival and legal technicality. acting_description: neutral, understated, quiet speed: 0.95 trailing_silence: 0.45 We are looking at a fundamental shift in the definition of operational drift. Usually, we talk about models drifting from their training data or their stated goals. Here, the drift is in the regulatory framework itself. The guardrails intended to protect the consumer from corporate monopolies are now being framed as the very things preventing the protection of the public from runaway technology. The mechanism of protection is being turned inside out. acting_description: sober, grounded, restrained speed: 0.94 trailing_silence: 0.5 The record shows that this idea has already begun to permeate the highest levels of the industry. Reports indicate that Elon Musk told Demis Hassabis of Google DeepMind that he wants leading artificial intelligence companies to have a regular call to discuss safety and security issues. Commentators like Dave Kasten have echoed this, suggesting that while the actual antitrust risk might be low, the government should explicitly waive it to remove the chilling effect on safety cooperation and allow for direct executive alignment. acting_description: factual, authoritative, precise speed: 0.96 trailing_silence: 0.4 In twenty-twenty-four, a government memorandum reportedly included a waiver process that allowed agency heads to grant limited exceptions, capped at one year, where relationships are necessary to responsibly steward United States national security. However, this was aimed primarily at government-to-lab relationships—procurement and threat intelligence—rather than the direct lab-to-lab coordination M

Read disclosure