This essay began as a response to claims that the argument that Chevron encourages congressional inaction has been refuted by the best available evidence. That Chevron causes such inaction is one of the arguments made by petitioners in Loper Bright. Leading scholars reject the argument. For instance, Chris Walker has called it one of the “least persuasive points” made at oral argument and Jim Speta has called it “empirically untrue.” Last year, Nicholas Bednar had a very good essay at Notice & Comment arguing that whether Chevron in fact has this effect is an empirical question. He went on to review literature related to (his framing of) this question and concluded by finding that the claim is unsupported and suggesting that its proponents lack candor.

Read the full piece here.

Margrethe Vestager, the European Union’s commissioner for competition, posits that competition law has not addressed “the structural entrenchment of companies holding market power”, and that sweeping regulations like the European Union’s Digital Markets Act (DMA) are therefore justified (By invitation, June 3rd). She compares the case-by-case approach of competition enforcement to “playing a never-ending game of whack-a-mole”. However, enforcement is often slow and complex, especially in the kinds of “abuse of dominance” cases that have been brought against large online platforms. This deliberate pace is necessary, as the companies’ business models and the consequences of their behaviour are themselves complex.

Read the full piece here.

We, the undersigned organizations and individuals, are writing to express our serious concerns about SB 1047, the Safe and Secure Innovation for Frontier Artificial Intelligence Systems Act. We believe that the bill, as currently written, would have severe unintended consequences that could stifle innovation, harm California’s economy, and undermine America’s global leadership in AI.

Our main concerns with SB 1047 are as follows:

1. The application of the precautionary principle, codified as a “limited duty exemption,” would require developers to guarantee that their models cannot be misused for various harmful purposes, even before training begins. Given the general-purpose nature of AI technology, this is an unreasonable and impractical standard that could expose developers to criminal and civil liability for actions beyond their control.
2. The bill’s compliance requirements, including implementing safety guidance from multiple sources and paying fees to fund the Frontier Model Division, would be expensive and time-consuming for many AI companies. This could drive businesses out of California and discourage new startups from forming. Given California’s current budget deficit and the state’s reliance upon capital gains taxation, even a marginal shift of AI startups to other states could be deleterious to the state government’s fiscal position.
3. The bill’s definition of a “covered model”–models trained with more than 10^26 floating-point operations at a cost above $100 million-will create confusion, encourage an adversarial relationship between the Frontier Model Division and AI developers, and interfere with industry dynamics in unpredictable ways. First, it is not always straightforward to say what a training run for a model costs. Second, the Frontier Model Division will have an incentive to investigate AI companies’ finances and other records to ensure they are not training covered models, which will create another burden for developers. Finally, it penalizes companies based on the size of their investment in AI: if one company trains a model above the threshold, they will be regulated in perpetuity. Yet because compute costs fall rapidly, a competitor could train a model six months later and be subject to no regulation at all. This is nonsensical.
4. The bill’s combination of the precautionary principle and liability (both criminal and civil) is incompatible with the way open-source software has been developed and distributed for decades. While this bill would not ban any existing open-source model, it constitutes a gradual legislated phasing out of open-source AI near today’s frontier.

These restrictions on open-source AI models would undermine a key driver of innovation and collaboration in the field. The vast majority of stakeholders, including large tech companies, startups, the broader business community, academia, and civil society organizations like the Center for American Progress, have voiced support for open-source AI development. Open-source AI has also thus far played an essential role in interpretability and safety research; by limiting access to future open-source models, this bill could undermine progress in those vital fields.

We believe that SB 1047, if enacted in its current form, would have a chilling effect on AI research and development in California and potentially across the United States. It could slow down progress in a field that holds immense promise for advancing scientific understanding, improving medicine, and driving economic growth.

While we share the goal of ensuring that AI is developed and deployed responsibly, we urge you to reconsider the approach taken in SB 1047. The bill is also broadly inconsistent with the legislative direction suggested by the United States Senate’s Bipartisan Working Group on AI; if SB 1047 passes, California likely would be an unfortunate outlier in the broader context of American policy stances toward AI. In conclusion, we respectfully request that you either make substantial changes to SB 1047 to address the concerns outlined above or withdraw the bill entirely. We stand ready to work with you to find a path forward that promotes innovation while also ensuring the safe and responsible development of AI technology.

Sincerely,

Neil Chilson, Head of AI Policy,, Abundance Institute

Kristian Stout, Director of Innovation Policy, International Center for Law & Economics

Lisa B. Nelson, CEO, ALEC Action

Logan Kolas, Director of Technology Policy, American Consumer Institute

Daniel Castro, Director, Center for Data Innovation

Taylor Barkley, Director of Public Policy, Abundance Institute

Adam Thierer, Resident Senior Fellow, Technology & Innovation, R Street Institute

Vance Ginn, Ph.D., Former Chief Economist, White House Office of Management and Budget

Jessica Melugin, Director, Center for Technology and Innovation, Competitive Enterprise Institute

Nathan Leamer, Executive Director, Digital First Project

Abstract

Competition benefits consumers, not just through lower prices and better quality, but also by protecting them against fraudulent practices. Clear-eyed government regulation can promote competition and consumer protection by stomping out fraudulent and deceptive practices as well as facilitating the flow of accurate, easy-to-understand information. But what happens when the government is the source of bad information and uses that to promote specious claims that competition has “failed”? In the case of the Consumer Financial Protection Bureau, another round of misleading economic analysis is being used to justify further intrusions on market competition that could confuse consumers and lead to worse regulation.

Read the full piece here.

Abstract

A February report from the Committee on Digital Competition Law (CDCL) recommended special competition rules for digital markets in India. It was accompanied by a draft Digital Competition Act (DCA) that is virtually identical to the European Union’s Digital Markets Act (DMA). Since it entered into force early last month, the DMA has imposed strict preemptive rules on so-called digital “gatekeepers,” a cohort of mostly American tech giants like Google, Amazon, Apple, Meta, and Microsoft.

Read the full piece here.

In a recent article for ProMarket, Jonathan Masur and Eric Posner defend the legality of the Federal Trade Commission’s rule banning noncompetes. However, little of their argument addresses the widespread contrary arguments against the ban.

On the Yale Journal of Regulation Notice and Comment Blog, Dan Crane recently shared the results of an informal survey of antitrust and administrative law professors. Of his 17 respondents, “Only one person predicted that the rule will be upheld.” Both the Wall Street Journal and Washington Post editorial boards have argued that the FTC lacks congressional authority to issue the rule. While it is risky to draw inference from silence, it seems notable that the New York Times editorial board, a long-time advocate for FTC Chair Lina Khan and early supporter of the noncompete rule, has not opined. A recent Congressional Research Service report characterizes the FTC’s legal authority to issue such rules as “unsettled.”

Read the full piece here.

Abstract

Scholars have widely criticized the Supreme Court’s Fourth Amendment jurisprudence as incoherent, especially in cases involving emerging technologies. This Article argues that to understand Fourth Amendment doctrine, one must consider how the values that underlie the Court’s decisions are balanced against each other and shift over time. To do so, this Article first proposes a novel, bottom-up approach to identifying the relevant values that focuses on the specific evidence that the Court considers in each case. Distilling the values underlying the Fourth Amendment provides a more coherent understanding of Fourth Amendment doctrine. This Article then applies this framework to three biometric technologies: facial recognition, iris recognition, and DNA profiling. Law enforcement use of these technologies may all raise Fourth Amendment challenges, but the framework shows how these challenges implicate different values. Recognition and application of this framework can result in a better appreciation of the impact of emerging technologies on Fourth Amendment doctrine.

Read at SSRN.