Pichai: DOJ Data-License Mandate is a De Facto Search Spin-Off
Background: DOJ’s Antitrust Case Against Google
On April 30, 2025, Google CEO Sundar Pichai took the witness stand in Washington, D.C., to counter the U.S. Department of Justice’s (DOJ) proposal to impose structural and data-sharing remedies on Google Search. The DOJ, which rested its liability case earlier in the week, accuses Google of maintaining an illegal monopoly in general search and search advertising. Its proposed solution would force Google to license its search index, user data, and certain proprietary ranking algorithms—measures that Pichai argues would amount to a “de facto divestiture” of Google’s core search business.
Technical Implications of Forcing Data Licensing
At the heart of the dispute is Google’s vast, continuously updated web index and the machine-learned ranking models that power it. Google’s index currently spans tens of billions of web pages, refreshed every millisecond via distributed crawler fleets running on Google Cloud’s custom Tensor Processing Units (TPUs) and high-throughput networking fabrics. The DOJ’s remedy would require Google to provide third parties API access to:
- The full keyword-searchable index, including metadata scores and freshness metrics.
- Historical click-through and query-behavioral datasets aggregated at the session level.
- Proprietary ranking signals drawn from real-time user engagement and machine-learning inference logs.
Experts warn that exposing these systems—even behind pay-per-call APIs—would enable competitors to replicate critical elements of Google’s engineering stack, effectively reverse-engineering its quality-of-service optimizations. Dr. Karen Liu, a former lead engineer on Google’s Search Infrastructure team, testified that such data sharing would undermine the privacy-preserving techniques built into Google’s federated analytics and differential privacy pipelines.
Innovation and R&D Under Threat
Pichai emphasized that Google’s multi-decade investment in R&D—totaling over $250 billion since 2004—funds projects ranging from next-generation natural language understanding in Bard and Gemini to quantum-aware search algorithms in Google Research’s AI Quantum lab. He argued that if rivals could license the same data and models for a nominal fee, Google would lose its incentive to build new infrastructure, platforms, and specialized accelerators.
“By forcing Google to share its deep index and model weights,” Pichai said, “the DOJ would effectively remove the feedback loop that drives our neural ranking improvements and latency optimizations, making continued investment unviable.” Independent analysts at IDC estimate that Google’s search-related R&D investment supports over 30,000 specialized jobs worldwide and underpins more than $100 billion in annual economic activity across the digital ecosystem.
Comparison with EU Digital Markets Act Precedents
The DOJ’s proposed data remedies bear resemblance to the European Union’s Digital Markets Act (DMA), which mandates so-called “gatekeeper” platforms to share certain interoperability and data access points with rivals. However, unlike the DMA—which specifically targets messaging and app stores—the DOJ’s plan is broader in scope, potentially covering Google Search’s entire backend. Legal experts caution that transplanting the DMA’s principles into U.S. law without guardrails could:
- Compromise user privacy by widening data exposure beyond Google’s tightly controlled environments.
- Create technical debt as multiple licensees integrate Google’s evolving index via stale APIs.
- Stifle cross-platform innovation by entrenching dependences on licensed assets rather than encouraging independent index creation.
Google’s Browser and AI Integration Strategy
In his testimony, Pichai also outlined Google’s ongoing efforts to maintain a healthy web ecosystem through open-source contributions. He pointed to Chromium—the open-source project powering Chrome—which currently holds approximately 65 percent of the desktop browser market. Google warns that divestment of Chrome or forced transfers to third parties could degrade security patch distribution and hamper the adoption of emerging web standards such as WebGPU and WebAuthn.
On the AI front, Pichai reaffirmed Google’s near-term agreement with Apple to integrate Gemini into Apple Intelligence on iOS. Slated for rollout later in 2025, this partnership would see Gemini’s large multimodal foundation models hosted on Google’s Vertex AI infrastructure, with selective compute tasks executed on-device via Apple’s Neural Engine. Unlike Google’s exclusive search arrangement—which the DOJ identifies as anticompetitive—this AI alliance is structured to let Apple add models from other vendors through standardized ONNX runtime interfaces.
Financial Performance Amid Legal Uncertainty
Despite ongoing legal challenges, Google’s financial results remain robust. In Q1 2025, Alphabet reported:
- Revenue: $90.23 billion, up 12 percent year-over-year.
- Net Income: $34.54 billion, a 46 percent increase over Q1 2024.
- Cloud Segment Growth: 28 percent annual growth, driven by BigQuery ML and Vertex AI adoption.
Analysts at Morgan Stanley recently raised Alphabet’s price target, citing strong generative AI product pipelines and accelerating demand for cloud-based AI services as offsetting near-term regulatory headwinds.
Looking Ahead: Potential Remedies and Appeals
Trial Judge Amit Mehta is expected to issue his ruling on remedies by August 2025. While he has expressed skepticism about divesting Chrome, he has already found Google liable for monopolistic conduct in the search market. Google has indicated it will appeal any final decree, seeking a stay on the implementation of remedies while the appellate process unfolds. The case’s outcome could reshape the competitive landscape of search, browser platforms, and AI services for years to come.