On September 2, the US District Judge Amit Mehta delivered a judgment that was possibly predictable but disappointing. While maintaining his August 2024 finding that Google illegally monopolised the search market, Mehta rejected the Department of Justice’s request to force Google to divest Chrome, calling such a breakup a “poor fit.” Instead, he opted for what amounts to a regulatory slap on the wrist: Barring Google’s exclusive search deals and requiring limited data sharing.
This outcome represents a fundamental misunderstanding of how monopoly power operates in the digital age, particularly as artificial intelligence transforms the very nature of search itself. The judge’s reluctance to pursue structural remedies reflects an outdated approach to antitrust enforcement that fails to grasp the interconnected nature of Google’s ecosystem. Chrome isn’t merely a browser — it’s the gateway through which Google collects invaluable user data that feeds directly into its search algorithms and advertising systems. By allowing Google to retain control over this critical infrastructure while merely prohibiting certain contractual arrangements, Mehta has essentially left the monopoly intact while creating the illusion of meaningful reform.
To understand the significance of this misstep, we must revisit Mehta’s landmark August decision. In his 277-page ruling, the judge definitively concluded that “Google is a monopolist, and it has acted as one to maintain its monopoly.” He found that Google violated Section 2 of the Sherman Antitrust Act through its web of exclusive deals with device manufacturers, browser makers, and wireless carriers — agreements that cost the company an estimated $26 billion annually but effectively locked out competitors from gaining meaningful market share.
That August ruling was hailed as the most significant antitrust decision since the Microsoft case two decades ago. It offered hope that courts were finally ready to confront the reality of digital monopolies and their anti-competitive effects. The decision recognised that Google’s dominance wasn’t simply the result of superior innovation, but rather the product of strategic barriers that prevented rivals from accessing the scale necessary to compete effectively in search.
Yet this week’s remedy ruling reveals the limitations of judicial thinking that remains anchored in industrial-age concepts of competition. Mehta’s solution — essentially forcing Google to compete for default search placements rather than securing them through exclusive contracts — treats the symptoms while ignoring the underlying disease.
The timing of this decision is particularly problematic given the seismic shift occurring in how we interact with information. The rise of large language models and AI-powered search tools like ChatGPT, Claude, and Google’s own Gemini represents the most significant transformation in information retrieval since the advent of web search itself. In this new landscape, the traditional boundaries between search engines, web browsers, and AI assistants are rapidly dissolving.
Google’s control over Chrome becomes even more strategic in an AI-driven world. Browser data provides crucial insights into user intent and behaviour patterns that are invaluable for training AI models. When users interact with AI-powered search features integrated into Chrome, Google gains access to a feedback loop that competitors simply cannot replicate. This data advantage becomes self-reinforcing, allowing Google to improve its AI capabilities while simultaneously making it harder for rivals to catch up.
The DOJ understood this dynamic, which is why they pushed for Chrome’s divestiture. A standalone Chrome would have levelled the playing field by preventing Google from leveraging browser data to strengthen its search and AI products. Instead, Mehta’s decision ensures that Google maintains this crucial advantage precisely when it matters the most.
The judge’s ruling on data sharing requirements further illustrates the inadequacy of the chosen remedies. While Google must now provide “certain search index data and user interaction data” to competitors, the court specifically excluded advertising data and limited the scope of shared information. This is akin to forcing an oil company to share some crude oil while allowing it to keep exclusive access to the most valuable refined products.
In the AI era, comprehensive data access isn’t just helpful — it’s essential for meaningful competition. Training effective AI models requires massive, diverse datasets that capture the full spectrum of user queries and interactions. By providing only limited data sharing, Mehta has created the appearance of promoting competition while ensuring that Google’s fundamental advantages remain intact.
The irony is palpable: Just as Microsoft’s bundling of Internet Explorer with Windows was seen as anti-competitive in the 1990s, Google’s integration of Chrome with its search and AI services represents an even more sophisticated form of tying. Yet where earlier courts were willing to consider structural remedies, today’s judiciary seems reluctant to disrupt established business models, even when they clearly stifle competition.
Clearly, the Department of Justice now faces a critical choice: They can appeal Mehta’s remedy decision, arguing that the prescribed fixes are insufficient to address the monopolisation the court itself acknowledged. Alternatively, they can accept this limited victory and focus their antitrust efforts on other aspects of Google’s business empire — the company still faces a separate case regarding its advertising technology practices.
However, this case reveals a broader problem with how antitrust enforcement approaches digital monopolies. The traditional toolkit of prohibiting certain contracts or requiring limited data sharing is simply inadequate for markets where network effects, data advantages, and ecosystem integration create nearly insurmountable barriers to entry.
Future antitrust enforcement must evolve to recognise that in digital markets, structural remedies aren’t just preferable — they’re often the only meaningful option. When companies control multiple layers of the technology stack, behavioural remedies rarely address the fundamental sources of their market power.
The Google case offers important lessons for how regulators should approach monopoly cases in the AI age. First, courts must better understand the interconnected nature of digital ecosystems and how control over one component can reinforce dominance across multiple markets. Second, remedies must be forward-looking, considering not just current market dynamics but how emerging technologies might alter competitive landscapes.
Most critically, antitrust enforcement must move beyond the consumer welfare standard’s narrow focus on prices and output. In digital markets, the harms from monopolisation often manifest as reduced innovation, decreased privacy protection, and the stifling of new business models that consumers don’t yet know they want.
The AI revolution offers both an opportunity and a challenge for competition policy. It’s an opportunity because new technologies can potentially disrupt established monopolies — we’ve already seen how ChatGPT has forced Google to accelerate its AI development. But it’s also a challenge because companies with existing data and computational advantages may be able to extend their dominance into new domains.
The writer is defence and tech policy adviser and former country head of General Dynamics
