FTC Antitrust Review of Tech M&A: What Triggers Scrutiny of AI Acquisitions

Last updated 4 months ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.

Meta reports earnings on January 28, 2026, with Wall Street braced for a number that has nothing to do with quarterly revenue: spending on equipment and infrastructure. When does this much investment in AI infrastructure cross from competitive investment into blocking competitors from the market?

The Federal Trade Commission is asking that question through its recent actions, suggesting it’s watching how dominant technology firms use money, talent acquisition, and partnership structures to concentrate power in artificial intelligence markets where traditional merger review tools weren’t designed to operate.

The FTC can challenge acquisitions and review deals through reporting requirements and past court decisions. What it doesn’t have is a clear framework for addressing competitive harms that don’t look like traditional mergers—when a big tech company hires an entire startup’s team while technically not buying the company, or spending so massive that smaller competitors can’t afford to compete, or cloud partnerships structured to avoid merger filing requirements while locking an AI developer into exclusive dependence on one provider’s computing resources.

Meta’s position illustrates the regulatory puzzle. The company hasn’t announced major AI acquisitions comparable to Microsoft’s hiring of Inflection AI’s 70-person team or Amazon’s recruitment of 80 percent of Adept’s technical staff. Instead, Meta is building through proprietary infrastructure that, once operational, could make external AI partnerships unnecessary.

Why Current Rules Don’t Capture These Deals

Federal law requires companies to report proposed mergers and acquisitions when the transaction size exceeds specific dollar thresholds. These rules were designed before artificial intelligence existed as a commercial market, before “talent” became the primary asset being acquired in technology transactions, before cloud computing created dependencies that function economically like exclusive dealing arrangements without involving any transfer of ownership.

The gap between what Hart-Scott-Rodino reporting captures and what reshapes AI competition has become a problem the FTC can’t ignore. The resulting staff report, published in early 2025, documented that major cloud providers had structured partnerships to include “significant equity and certain revenue-sharing rights,” exclusive consultation provisions, requirements that AI developers spend large portions of funding on the cloud provider’s services, and access to sensitive technical information unavailable to competitors.

None of these arrangements necessarily triggered reporting requirements. Many involved minority stakes below the threshold, or licensing agreements paired with talent recruitment, or infrastructure commitments structured as service contracts rather than acquisitions. Yet each arrangement concentrated market power.

What Triggers FTC Review

The 2023 Merger Guidelines establish analytical frameworks for assessing whether a merger raises competitive concerns. The guidelines explicitly address when a big company buys a small startup that could become a major competitor—situations where a dominant firm acquires a startup that hasn’t yet commercialized products but possesses clear potential to disrupt existing markets.

Under this doctrine, the FTC will challenge an acquisition when a dominant firm buys a startup with potential to enter the dominant firm’s market, even when traditional market concentration metrics don’t suggest harm. The question is whether the startup possessed capabilities, resources, or talent positioning it as a likely future entrant.

For Meta, this doctrine creates exposure. The company’s position in social networking remains formidable. The district court’s finding that Meta holds no monopoly creates a significant barrier to the appeal. However, the decision to pursue the appeal anyway—announced by Bureau of Competition Director Daniel Guarnera with the statement that “the Trump-Vance FTC will continue fighting its historic case against Meta”—signals the agency views the case as establishing precedent about dominant firms’ acquisition strategies. This suggests the agency distinguishes between promoting AI innovation generally and preventing dominant firms from using acquisitions to suppress competition.

How Companies Avoid Deal Reviews by Hiring Teams Instead

Hiring arrangements exemplify what the FTC views as potential ways to avoid government approval requirements. The legal challenge is distinguishing between normal employment and anticompetitive talent acquisition. When a dominant firm hires talented engineers from a startup, is that competitive behavior or blocking competitors from accessing resources?

Hiring talented engineers is ordinarily pro-competitive. Competition for talent should drive higher wages and better working conditions. The concern isn’t that Microsoft is hiring talented people—that’s competition for talent. The concern is that by hiring an entire team from a startup, dominant firms eliminate the startup as an independent competitive threat while gaining access to the startup’s technology and talent.

U.S. law distinguishes between whether the deal hurt the whole market or just one company. Hiring talented employees, even in large numbers, appears to harm Inflection as a competitor but might benefit competition if Microsoft’s AI systems improve as a result. The FTC must demonstrate that the hiring reduced competition in AI development generally, not merely that it harmed Inflection’s prospects.

That requires showing there would have been meaningful competition between Microsoft and Inflection absent the deal, and that eliminating this competition reduced innovation or consumer welfare in AI markets. Given that Inflection was relatively young with no substantial customer base or revenue, this showing may be difficult.

Meta hasn’t announced hiring transactions comparable to Microsoft, Amazon, and Google’s 2024 deals. The company has expanded its AI engineering workforce through organic hiring and some talent acquisition, but not the wholesale team recruitment that triggered regulatory scrutiny of competitors. Should Meta pursue similar strategies, expect FTC examination.

Building an Unbeatable Cost Advantage

Meta’s investment strategy can be understood as an attempt to compete with Microsoft, Amazon, and Google by building internal capabilities to render external AI startups unnecessary. If Meta can develop AI systems Meta owns and controls—Llama, and the new Avocado model scheduled to debut in early 2026 to “compete with Gemini and ChatGPT”—that serve the company’s social media and advertising use cases, then Meta captures value that would otherwise flow to external AI startups or other cloud providers.

This strategy requires extraordinary investment in the near term. Training advanced large language models requires massive computing capacity that only the wealthiest technology companies can afford. Meta’s development of Llama 2 required processing “billions of data points daily,” with over $40 billion invested in R&D to date, a significant portion allocated to AI and machine learning.

OpenAI and other AI startups have survived only through securing funds from technology giants. Microsoft’s cumulative investment in OpenAI reached $13 billion as of early 2025. Amazon committed $4 billion to Anthropic. Google made parallel investments in multiple AI ventures. These partnerships enable AI startups to scale but simultaneously create dependencies and lock-in effects that constrain competitive autonomy.

An AI startup using Microsoft’s Azure cloud to train and deploy models faces substantial costs if it later wants to migrate to alternative providers or develop independent resources. Microsoft’s exclusive rights to use OpenAI’s technology, granted in their partnership agreement, further constrain the startup’s ability to compete independently or serve Microsoft’s competitors.

The 2025 staff report documented that cloud service providers structured agreements requiring AI developers to “spend a large portion of their cloud service provider partner’s investment on cloud services from their partner” and secured “significant equity and certain revenue-sharing rights.” These terms mean that when a startup raises funds from Microsoft, Amazon, or Google, the funds are economically constrained to benefit the provider’s cloud services business.

Could the FTC challenge Meta’s investment as anticompetitive? Theoretically, if Meta’s investment creates a situation where alternative AI developers cannot access sufficient computing resources at competitive prices due to Meta’s consumption of capacity, the agency might view such investment as blocking competitors from getting computing power.

But such a challenge would represent extraordinarily aggressive action, stretching FTC authority into territory where it has never operated. The agency has never successfully challenged a firm’s expenditure on resources as anticompetitive. Courts have historically been reluctant to second-guess business investment decisions. The doctrine would require proving that Meta’s investment was undertaken primarily to block competitors’ access to computing resources rather than to serve Meta’s operational needs and business strategy.

Given the Trump administration’s stated commitment to fostering AI innovation and supporting U.S. technology leadership in global competition with China, such a challenge appears extraordinarily unlikely under current regulatory leadership.

FTC Leadership and Enforcement Direction

The transition from Lina Khan to Andrew Ferguson as FTC Chair represents a change in how strictly the FTC enforces competition rules. Khan, serving from 2021 through the 2024 presidential transition, pursued what she characterized as a “modern” approach focused on preventing dominant firms from using acquisitions to suppress competition in emerging markets. This led to challenges to Microsoft’s Activision Blizzard acquisition, investigation into the Microsoft-OpenAI partnership, and appeal of the Meta monopolization case that resulted in the November 2025 district court loss.

Ferguson came to the position with a track record of disagreeing with actions Khan pursued. Most significantly, Ferguson dissented from the FTC’s 2023 nationwide ban on non-compete agreements, arguing the agency went beyond what the law allows under the FTC Act’s prohibition on “unfair methods of competition.”

Yet Ferguson’s statements regarding AI suggest the new leadership hasn’t abandoned merger scrutiny of tech companies or AI transactions. In early 2026 remarks, Ferguson stated the agency would “continue to serve as a vigilant competition watchman,” particularly in “rapidly evolving sectors like artificial intelligence.”

The appointment of Mark Meador as the fifth FTC Commissioner complicates the picture. Meador, like Khan, has been “outspoken in his critiques of Big Tech” and has proposed that actions should focus on preventing anticompetitive behavior. His presence suggests technology merger review won’t revert to the permissive stance of earlier administrations but will pursue what Ferguson describes as “carefully evaluated” actions focusing on cases where likelihood of success justifies deployment of limited resources.

The Trump administration’s July 2025 AI Action Plan explicitly committed to reviewing prior investigations and orders “to ensure none advance theories of liability that unduly burden AI innovation.” This directive resulted in the December 2025 decision to set aside its 2024 final settlement agreement against Rytr, an AI writing service, finding the order “unduly burdens artificial intelligence innovation.”

The concurrent appeal of the Meta monopoly case suggests the agency distinguishes between actions against novel AI companies like Rytr and actions against dominant incumbent firms using acquisitions to maintain market dominance. The statement in the Rytr decision that “the FTC will continue to hold accountable actors that use AI to violate the law or deceive consumers” signals the agency is distinguishing between regulating AI innovation generally and preventing deceptive practices in AI markets.

How Meta’s Strategy Compares to Competitors

Microsoft secured exclusive rights to use OpenAI’s technology while maintaining significant equity stakes, effectively making OpenAI’s models central to Microsoft’s Azure cloud business. This grants Microsoft privileged access to advanced AI technology, which it integrates into Office productivity suite, Azure cloud services, and Windows operating system, using its existing customer base to sell AI products across multiple customer segments.

Google pursued a more diversified approach, investing in both internal AI development (Gemini and Bard models) and external partnerships with AI startups through Google Ventures and direct investment arms. The company has acquired AI-focused startups and technical talent, though generally at smaller scale than headline acquisitions by Microsoft, Amazon, and Alphabet in other technology domains.

Amazon structured its AI strategy around AWS cloud services, integrating AI capabilities into existing cloud offerings while simultaneously investing in AI startups and technology companies that could serve as future acquisition targets.

Nvidia occupies a unique position as hardware supplier to nearly all other AI players. With greater than 90 percent market share in specialized AI accelerator chips (GPUs), Nvidia has positioned itself as the input provider to all downstream AI development. Nvidia’s recent $100 billion investment in OpenAI raised concerns from regulators who view the arrangement as making Nvidia and OpenAI partners, which could create conflicts of interest. The investment signals Nvidia is expanding from pure hardware supplier toward investor and strategic partner in major AI systems, potentially raising concerns about whether Nvidia might favor OpenAI over competing AI developers in GPU allocation, pricing, or technical support.

In this context, Meta’s investment represents an attempt to reduce dependence on external AI partners by developing proprietary models and resources. Unlike Microsoft, Meta hasn’t secured exclusive partnerships with leading independent AI labs. Unlike Google, Meta isn’t leveraging a dominant search position to distribute AI products. Unlike Amazon, Meta isn’t leveraging a dominant cloud services position to integrate AI across existing customer relationships.

Instead, Meta is deploying funds directly with the goal of generating proprietary AI capabilities that Meta can monetize through advertising improvements, new platforms (WhatsApp Business, Threads), and potentially through licensing arrangements with third parties. This strategy requires demonstrating that investment generates tangible business returns—engagement improvements, ad targeting accuracy increases, and new revenue streams.

What to Watch in Meta’s January 28 Announcement

Analysts expect revenue of approximately $58.3 billion (reflecting about 20-21 percent year-over-year growth) and profit per share (adjusted for one-time costs) of $8.16. More significant is what executives say they plan to spend in 2026.

The regulatory trajectory for scrutiny of Meta’s AI strategy depends on several factors. First, whether Meta’s investment generates claimed competitive benefits in engagement, ad targeting, and new platform monetization will influence whether regulators view the investment as pro-competitive innovation or anticompetitive buying up computing power to keep competitors from using it. If Meta demonstrates that the Avocado large language model, debuting in early 2026, significantly improves ad targeting or user engagement, regulators may view investment as justified innovation. Should investment fail to generate proportionate returns, regulators might scrutinize whether Meta was creating barriers to entry that smaller AI companies cannot overcome.

Second, whether Meta pursues major AI acquisitions or hiring arrangements in 2026 will influence whether the FTC opens new actions. Meta has been less aggressive than Microsoft, Amazon, and Google in recruiting AI talent from startups or acquiring AI-focused companies. Should this pattern continue, attention may focus elsewhere. Should Meta shift strategy toward major acquisitions or talent-concentrated hiring arrangements, the agency would likely examine whether such deals raise concerns about startups that could become major competitors later comparable to those identified in prior Meta acquisitions.

Third, the composition and priorities of the FTC under Ferguson’s leadership will influence whether pending investigations into Big Tech AI practices are resolved through settlement or litigation. The Trump administration’s stated commitment to AI innovation may create political pressure to resolve investigations without imposing conditions viewed as unduly burdensome to AI development. Conversely, Ferguson’s stated commitment to remain “a vigilant competition watchman” suggests genuine competitive harms will continue to trigger agency action even if activity becomes more restrained than during the Khan administration.

The Regulatory Challenge: Can Antitrust Law Keep Up?

The FTC must apply competition law to AI markets. But traditional deal review rules may not catch the biggest competitive harms. Hiring arrangements, minority stake investments, and resource investment all can reduce competition while potentially evading federal deal approval requirements or traditional merger analysis frameworks.

For Meta specifically, the company faces potential scrutiny from multiple directions: continued appellate litigation regarding historical Instagram and WhatsApp acquisitions, the FTC could investigate any major AI deals Meta announces in 2026, and possible broader scrutiny of whether investment creates barriers to entry that harm AI competition.

The broader question driving regulatory attention to AI deals is whether competition policy can adapt quickly enough to address harms created by rapid technological change. Traditional merger review operates on timescales measured in months. Artificial intelligence technology evolves on timescales measured in weeks. Deciding whether a startup with no revenue but advanced AI research capabilities should be protected as a startup that could become a major competitor later requires predicting future technology development, market adoption, and competitive threats—predictions that could easily be wrong.

The honest answer is that the agency is still figuring out what triggers scrutiny of AI acquisitions. Formal thresholds exist. Doctrinal frameworks provide guidance. But the gap between what current law captures and what reshapes AI competition remains substantial. As the agency works through these challenges under evolving leadership with shifting philosophies, decisions made regarding Meta’s AI strategy may establish precedent influencing how regulators approach tech acquisitions for years to come.

The earnings call will provide some answers. The January 28 announcement will signal whether Meta’s bet is paying off, and whether the company’s approach to AI competition will invite the kind of regulatory scrutiny that has already ensnared its competitors.

Our articles make government information more accessible. Please consult a qualified professional for financial, legal, or health advice specific to your circumstances.