Too little, too late: The inadequacy of reactive antitrust enforcement in technology markets

After finding in April 2025 that Google violated antitrust law through control of the functional aspects of online advertising, Judge Leonie Brinkema recently issued her opinion as to the remedies to be imposed. The Justice Department and various state attorneys general had asked that the integrated components she found central to Google’s monopolization of the ad tech market be spun off to allow for a competitive market going forward. But Brinkema declined to order divestiture of any of those assets. Instead, she ordered a series of behavioral remedies intended to prevent continued abuse of Google’s monopoly power. 

This is the second time in little more than a year that a liability finding against Google was followed by a mild “go and sin no more” behavioral remedy. In the earlier case, Judge Amit Mehta found that Google unlawfully monopolized online search but declined the Justice Department and state attorneys general’s request to break up the monopoly.

Despite the good faith efforts of antitrust enforcers and the courts in both cases, the outcomes in each have sparked an intense and legitimate debate over whether such relatively modest behavioral remedies will prove sufficient to create a competitive market going forward and deter future exploitative behavior by these and other dominant tech firms. 

We need to reassess whether antitrust laws written in the industrial era are up to the challenge presented by the fast-paced digital era. After-the-fact enforcement in these markets suffers from a painfully slow judicial process that aids the alleged monopolist. This begins with how, absent early injunctive relief, monopoly abuse can continue for years before trial and a decision on the merits. During that process, the technology and marketplace dominance typically controlled by the abuser speed ahead. Responding to those advances, markets continue to default to the dominant position of the gatekeeper. By the time the matter finally reaches trial, both technology and its application have redefined reality beyond the issues being litigated, making meaningful remedies illusive.

This problem is only compounded by the breathtaking pace of artificial intelligence (AI) development and the risk that just a few firms will come to dominate the AI stack.

It appears the judges in both Google cases recognized how technology had outpaced their respective courtroom debates. Whether they feared judicial intrusion into the magic of digital technology or were hoping new advances would help solve the issues they were being asked to decide, we don’t know. Regardless, these decisions reinforce the need to rethink the antitrust model for modern times.

4 specific examples

Since 2020, federal and state enforcers have brought four antitrust actions alleging abusive behavior by Google, Meta, and Amazon. Each became caught up in the time warp of technology.

Google began paying Apple and others to be their exclusive default search engine as early as 2005. The Federal Trade Commission (FTC) investigated in 2011 but took no enforcement action. Nine years later, the Department of Justice (DOJ) and a coalition of state attorneys general filed suit. But it took another four years for Mehta to conclude, after a lengthy trial, that “Google is a monopolist and it has acted as one [for at least 14 years] to maintain its monopoly.” The controversial behavior remedies decision took another year. And, of course, both Google and the antitrust enforcers are appealing, while most of the key remedial provisions are yet to be implemented.

The challenge to Google’s dominance of ad tech followed a similar path: A lawsuit filed in 2023 challenging acquisitions and conduct beginning as early as 2008 resulted in findings last year that the company had indeed engaged in conduct that “substantially harmed Google’s publisher customers, the competitive process, and, ultimately, consumers of information on the open web.” Here we are 18 years later awaiting the implementation of the judge’s behavioral remedies and, of course, the inevitable appeals. 

The FTC’s 2020 challenge to certain Meta behaviors, especially its acquisition of would-be competitors in 2012 and 2014, also made it to trial. Five years later, the trial judge ruled in Meta’s favor without deciding whether the company had acted badly over a decade earlier. He avoided that issue by concluding that the market was sufficiently different today and accordingly there was no present-day monopoly abuse for him to remedy. 

The FTC’s 2023 challenge to Amazon’s alleged abuse of its online shopping dominance dating to 2011 is the one Big Tech platform case that remains unresolved, awaiting a trial next year. That’s some more than 15 years before we will learn whether Amazon abused its gatekeeper status to the benefit of its shareholders and to the detriment of sellers and consumers. 

Lessons learned

Our antitrust laws are intended to deter bad behavior. But when a company weighs the profits from extracting years—and we mean years—of monopoly overcharges against the risk that sometime down the road antitrust enforcers may investigate, may challenge the misconduct, and may prevail; a court may force a company to change its behaviors going forward, but not require divestitures; and ultimately enforcers may prevail on appeal, deterrence seems illusory at best. In such circumstances, antitrust risk becomes a mere cost of doing business, a balance sheet line item that is dwarfed by the profits to be made while the issues are litigated. 

As former enforcers and believers in the value of competitive markets, we need to rethink the antitrust paradigm so that competition can thrive in technology markets, especially in the brave new world of AI. Antitrust enforcers require tools to intervene before gatekeepers can indefinitely exploit a market that has tipped in their favor. And, in the case of AI, we also need an oversight authority with the ex ante power to ensure that all benefit from a market that is robustly competitive.