Big Tech Faces Pressure on All Fronts

A New Phase of Pressure on Big Tech

Technology companies are confronting a widening front of scrutiny, as officials in Washington and Brussels press them over risks that now span national security, market power and child safety.

In Washington this week, top U.S. economic officials reportedly met with major bank leaders to discuss a new kind of threat: whether advanced artificial intelligence systems could help hackers discover and exploit software vulnerabilities quickly enough to endanger critical financial infrastructure. At nearly the same moment, the Trump administration was intensifying its criticism of the European Union over a stream of multibillion-dollar penalties against American technology firms. And in Europe, some of the same companies under antitrust and regulatory pressure were denouncing lawmakers for allowing a legal basis for detecting online child sexual abuse material to expire.

Taken together, the episodes suggest that Big Tech is entering a more complicated era, one in which the industry is no longer being challenged on a single axis. The largest firms and newest A.I. developers are being treated, increasingly, as actors whose decisions can carry consequences for banking stability, trade relations and public safety.

Banks Are Warned About A.I.-Driven Cyber Risk

The most striking sign of that shift came in the private warnings delivered to the financial sector. Treasury Secretary Scott Bessent and Jerome Powell, the chair of the Federal Reserve, were reported to have convened leaders of major U.S. banks in Washington to discuss the cyber implications of Anthropic’s new Mythos model.

Anthropic has restricted access to Mythos to a small group of companies rather than releasing it broadly, an unusual move that reflects concern that the model’s capabilities could be misused. The company has said it has been briefing U.S. officials on the model’s cyber potential.

The worry is not simply that A.I. can automate routine coding tasks. Officials are increasingly focused on whether frontier systems might materially improve offensive cyber operations — helping attackers identify weaknesses in software, accelerate exploitation or scale attacks against essential systems. For banks, that possibility carries special weight, since financial institutions sit at the center of payment networks and other critical infrastructure whose disruption could spread quickly.

For years, cyber policy has largely centered on software flaws, ransomware gangs and state-backed intrusions. What is changing now is the possibility that advanced models could become force multipliers. If that fear hardens into consensus, it could push governments toward stricter expectations for how frontier A.I. systems are tested, who gets access to them and what safeguards developers must put in place before deployment.

It remains unclear how immediate the real-world danger from Mythos is. But the fact that senior U.S. officials are discussing it directly with Wall Street executives points to a notable change in posture: frontier A.I. is being approached not just as a driver of productivity or innovation, but as a possible source of systemic risk.

Europe’s Enforcement Campaign Deepens a Rift

At the same time, long-running European efforts to curb the power of major American technology companies are becoming more entangled with geopolitics.

The latest flashpoint followed the European Commission’s €2.95 billion fine against Google in September over ad-technology practices, one of the bloc’s largest recent penalties. That case added to more than $7 billion in fines imposed on major U.S. tech companies in roughly the past two years, according to figures cited by administration officials and industry critics. The scale of those penalties has fueled fresh anger in Washington, where President Trump and his allies have portrayed Brussels’s campaign as discriminatory toward American firms.

The clash reflects more than irritation over individual cases. Europe has spent years building a broad architecture for digital regulation, combining traditional antitrust enforcement with newer rules under the Digital Markets Act and Digital Services Act. Brussels has argued that large platforms act as gatekeepers that distort competition and require stricter obligations. American officials, by contrast, have increasingly cast those actions as economic pressure aimed disproportionately at U.S. champions.

That dispute matters because tech policy no longer sits apart from trade diplomacy. Washington has openly suggested that foreign digital penalties and rules could provoke tariff threats or other retaliatory steps. If Europe continues to pursue large fines and structural remedies, the confrontation could widen beyond regulatory complaints into a more formal transatlantic economic conflict.

Child-Safety Rules Expose Another European Fault Line

Yet Europe is also facing pressure in the opposite direction — not for regulating too aggressively, but for failing to preserve a legal tool that major platforms say is essential for finding child sexual abuse material.

On April 3, a temporary exemption from the bloc’s privacy rules expired after the European Parliament did not extend it. The carve-out, first introduced in 2021, had allowed companies to voluntarily scan communications for child sexual abuse material, grooming and related harms while lawmakers worked on a permanent framework. It had already been extended once, in 2024, but only through this month.

After the lapse, Google, Meta, Microsoft and Snap sharply criticized the decision, warning that the loss of legal cover for automated detection could lead to fewer reports of abuse and more undetected crimes. Child-safety experts have voiced similar fears, pointing to earlier periods of legal uncertainty that led to significant declines in reporting.

The impasse reflects a familiar European struggle to reconcile two powerful principles: privacy protections, including concerns about scanning private or encrypted communications, and the demand to give platforms workable authority to detect and report exploitation. Lawmakers have yet to find a compromise that satisfies both camps.

For the companies involved, the moment is notable. Many of the same firms that have sparred with Brussels over competition and platform rules are now urging the bloc to restore a legal basis for enforcement on safety grounds. That inversion underscores how deeply technology platforms are embedded in public functions that governments increasingly expect them to help carry out.

Why This Moment Matters

What links these disputes is a broader redefinition of the technology sector’s role.

A.I. developers are no longer seen only as creators of tools for consumers and businesses; they are also being scrutinized as potential amplifiers of cyber capability. Established platforms are no longer facing just antitrust complaints; they are being asked simultaneously to comply with competition mandates, defend children from online exploitation and navigate conflicting demands over privacy. And governments are no longer treating tech oversight as a domestic matter alone; digital regulation has become part of a larger contest over trade, sovereignty and strategic power.

For banks and other critical industries, the Anthropic episode may prove especially consequential if it prompts formal standards around frontier-model testing, red-teaming and controlled release. For Europe and the United States, the regulatory fights are becoming a measure of how far allies can diverge on digital governance before those disagreements spill into economic retaliation.

The result is a sector under pressure from almost every direction at once — from regulators seeking to curb dominance, from officials trying to anticipate new security threats and from lawmakers struggling to define how much responsibility platforms should bear for harms that unfold on their services.