Courts and enforcers must not be shy about stopping harmful deals.
By Dan Mogin, Tim LaComb, and Joy Sidhwa
AI-related deal activity so far this year looks less like ordinary market consolidation and more like a race to secure long-term dominance.
Transactions increasingly cluster around firms that already control critical layers of digital infrastructure, raising an important question: what role will antitrust play when market power is assembled not through a single transformative merger, but through a dense web of acquisitions, investments, and strategic infrastructure commitments?
Public attention has largely focused on whether artificial intelligence itself should be regulated. But as Dan Mogin argued in his March 31, 2026 article, “Why AI Regulation Fails at the Enforcement Stage,” many such efforts are likely to falter because regulators cannot meaningfully observe, monitor, or control a fast-moving, widely dispersed general-purpose technology through prescriptive rules alone.
That critique matters here. Whatever happens on the regulatory front, competitive conditions are being reshaped now through private transactions that secure control over computing, infrastructure, and distribution channels—often long before significant regulatory frameworks take shape.
AI Deal Volume and the Reshaping of Market Structure
Analysis of AI‑related transactions reveals a striking pattern. More than 200 significant deals tracked in the Mogin Law A.I. Deal Table™ span acquisitions, equity investments, partnerships, and major financing rounds dating back to 2012, with recent activity showing a marked shift toward long‑duration infrastructure arrangements for cloud capacity and compute power (e.g., Meta-Nebius, AWS-Anthropic, Oracle-OpenAI).
These transactions range from nine‑figure agreements to mega‑deals in the tens of billions of dollars, reflecting a focus not merely on products, but on control over essential inputs.
Crucially, many deals do not simply acquire existing capacity. They pre‑allocate future access to electricity generation, AI‑ready data centers, advanced chips, and high‑capacity data movement—resources that economists and competition authorities have identified as foundational to participation in AI markets.
While no single aggregate figure captures the full scope of this activity, the size and frequency of these commitments suggest capital flows that, over time, reach well into hundreds of billions of dollars and beyond.
Market structure is being influenced not only by who owns which assets today, but by who has secured priority access to the resources that will determine participation tomorrow.
Today’s deal activity is shaping competitive conditions before traditional concentration metrics can even register meaningful change. Market structure is being influenced not only by who owns which assets today, but by who has secured priority access to the resources that will determine participation tomorrow.
Infrastructure Control, Chokepoints, and Competitive Dependency
Generative AI helps explain the urgency behind this consolidation. Unlike earlier technological advances that primarily automated discrete tasks, generative AI produces functional substitutes for human cognitive output across a wide range of activities, from coding and analysis to design and prediction.
These systems are increasingly used to accelerate their own deployment, supporting software development, market analysis, and operational decision‑making that facilitate further AI integration.
In this environment, early control over infrastructure and distribution channels can translate into a lasting advantage. Deal activity therefore gravitates toward chokepoints in the AI technology stack rather than consumer‑facing applications alone.
Viewed individually, many of these arrangements appear incremental or even efficiency‑enhancing. Viewed collectively, they can determine who is able to compete, innovate, or scale independently.
For firms operating adjacent to dominant platforms, the consequences may emerge quietly.
Competitive harm need not take the form of immediate price increases or outright exclusion. Instead, firms may find their strategic autonomy constrained as access to computing, data, or deployment pathways becomes increasingly mediated by counterparties with entrenched positions.
Dependency replaces rivalry not by design, but by accumulation.
Antitrust Law as a Check on Cumulative AI Consolidation
These dynamics place antitrust law in a central, if challenging, position. Antitrust is not well‑suited to regulating technology itself, particularly when markets evolve faster than enforcement can collect itself. But it remains the primary legal framework for evaluating how control over critical inputs and platforms is acquired and maintained.
The difficulty lies in timing and aggregation. Traditional antitrust analysis often evaluates transactions in isolation, assessing their effects within defined markets.
Generative AI complicates that approach. Market boundaries are fluid, competitive effects may be delayed, and structural change can occur through the interaction of many deals rather than a single transaction.
As competition authorities have increasingly recognized in scrutinizing cloud infrastructure, vertical integration, and interoperability, the risk is not merely higher prices, but the quiet foreclosure of future competition.
By the time conventional indicators register harm, competitive pathways may already be closed.
Courts and enforcers must stifle mergers that substantially lessen competition prospectively. Otherwise, damage can be done before realistic remedies can address the wreckage.
In the age of generative AI, antitrust law’s enduring task is not to slow innovation, but to ensure that innovation remains contestable.
As firms race to secure long‑term advantage through infrastructure control and cumulative acquisitions, the law’s relevance may lie less in regulating AI’s capabilities than in preserving the conditions under which new entrants—and new ideas—can still emerge.
Courts and enforcers must not be overly cautious in applying the Clayton Act’s directive to stifle mergers that “substantially lessen competition” prospectively.
Otherwise, as recently observed in the other aspects of Big Tech, the situation may rapidly spiral out of control with damage being inflicted before legal recognition of these issues and before realistic remedies can address the resulting wreckage.
Dan Mogin is Managing Partner at Mogin Law LLP. Tim LaComb and Joy Sidhwa are both Senior Counsel at the firm. All three professionals focus their practices on antitrust and competition law, advising and litigating on behalf of clients. Contact them at Info@MoginLawLLP.com.
What is the lesson ….?
The race for dominance in the development and application of generative AI tools could hobble the ability of startups and mid-sized innovators to compete. Attorneys at Mogin Law LLP are available to answer questions and discuss your concerns.
Takeaways
- AI dealmaking increasingly targets infrastructure rather than applications, with long duration commitments reshaping market structure before competitive balance can form.
- The scale of capital flowing into AI transactions is extraordinary, with individual deals reaching tens of billions of dollars and cumulative commitments reshaping competitive dynamics over time.
- Antitrust law may serve as the primary check on AI‑driven consolidation, not by regulating the technology itself, but by policing how control over critical inputs and chokepoints is assembled.
- Courts and enforcement agencies must tackle mergers that are likely to harm future competition; otherwise, the resulting damage could be irreversible.
If you have questions contact us at Info@MoginLawLLP.com.
Members of the press are encouraged to contact us at Media@MoginLawLLP.com.
FAQ
You should contact a qualified competition law firm that has experience with these types of matters. Getting a high-level assessment of whether you have a potentially actionable antitrust case is a good business practice. Contact Mogin Law LLP if you would like to schedule a consultation.