Google is positioning itself not just as an AI model provider, but as an infrastructure layer that connects models, tools, and workflows across enterprises. Two major announcements reveal this shift: Gemini Enterprise for Legal, developed in partnership with leading law firm Weil, and a rebuilt Google Ads API Developer Assistant that now functions as a global plugin across multiple coding environments.

What Is Google's New Legal AI Platform, and Why Did a Top Law Firm Help Build It?

Google launched Gemini Enterprise for Legal as a comprehensive offering that goes far beyond language models. The platform includes models, interfaces, agents, plugins, and supporting infrastructure designed specifically for legal workflows. Weil Gotshal & Manges, one of the leading US law firms, was instrumental in shaping the product from the ground up.

"The Gemini Enterprise for Legal thing specifically is the way Google is going to market with its models and what I'll call the supporting infrastructure for them. So it can be the models, it can be the way you interface with the models, whether through the web or through the anti-gravity at this point, two point zero or IDE application that sits on the desktop," said Andrew Simon, Chief AI and Innovation Officer at Weil.

Andrew Simon, Chief AI and Innovation Officer at Weil Gotshal & Manges

Weil contributed two early use cases to the platform's development: parallel research agents that allow lawyers to conduct multiple research lanes simultaneously on similar topics, and NDA drafting capabilities that enable agentic document generation from simple to increasingly complex agreements. The firm is not stopping there; more applications are expected to follow as the partnership deepens.

What makes this arrangement significant is that Weil helped shape the product while also planning to use it internally. The firm is building its own applications on top of the Gemini Enterprise Suite and Google Cloud infrastructure, including BenchMark, a system for analyzing judges' records to help litigators prepare cases more effectively.

How Does Google's Approach Solve the "45 Different Places" Problem for Law Firms?

One of the biggest challenges in modern legal tech is fragmentation. Lawyers need access to specialized tools like Harvey, Legora, and Thomson Reuters, but jumping between multiple platforms creates friction and slows down work. Google's strategy is to act as a centralized integration layer that brings these tools together without forcing lawyers to abandon their existing workflows.

Weil's approach demonstrates this philosophy in action. While the firm is not currently part of the Google Workspace ecosystem, it is testing integration. The key insight is that Microsoft Word does not need to be where all the computational work happens; it can simply be the surface where lawyers review final work products.

"What this moment is allowing us to do is decide: Is Microsoft Word the surface? Possibly, very likely, right? Everybody's on it. But it doesn't mean that needs to be where all of the work crunching happens," explained Andrew Simon.

Andrew Simon, Chief AI and Innovation Officer at Weil Gotshal & Manges

This flexibility enables a hybrid architecture where Google Cloud infrastructure handles model inference, document processing, and data retrieval in the background, while lawyers continue working in the familiar Word interface with lightweight plugins for interaction. The goal is to avoid forcing lawyers to jump between what Simon described as "45 different places".

What Strategic Advantages Does Google Offer Beyond Just Models?

Google's value proposition extends well beyond providing language models. The company offers what Weil calls the "whole vertical slice": data storage, retrieval, processing, model deployment, and iteration capabilities all on one infrastructure. This matters because it enables law firms to build mini-systems quickly, experiment with new applications, and scale what clients actually want without managing multiple vendor relationships.

Critically, Google does not lock firms into using only Google models. Weil can deploy Google models alongside OpenAI, Anthropic's Claude, and local models, all running on the same infrastructure layer. This multi-model strategy gives firms flexibility while centralizing the underlying compute and data management.

How Is Google Rebuilding Its Ads API Tools for AI-Driven Development?

Google's approach to enterprise infrastructure extends beyond legal tech. The company released version 4.0.0 of its Google Ads API Developer Assistant, a major architectural rebuild that transforms the tool from a standalone local workspace into a globally available plugin. This shift reflects a broader trend in how enterprises are adopting AI coding tools.

The new plugin model is designed to work across multiple development environments and projects, rather than requiring separate installations for each code repository. Importantly, the Assistant now supports both Antigravity and Anthropic's Claude Code, giving development teams flexibility in which AI coding environment they standardize on.

What Practical Features Make the New Ads API Assistant Different?

The rebuilt Assistant moves beyond simple code suggestions into validation, troubleshooting, and real-time reporting. Key features include:

• GAQL Validation: The Assistant automatically validates Google Ads Query Language syntax, field support, date segmentation, and zero-impression constraints before queries execute, preventing bad queries from reaching production and breaking dashboards or budget pacing alerts.
• Schema Inspection: The tool can read Protobuf schemas to identify available fields, data types, and enum definitions tied to specific Google Ads API versions, preventing the common mistake of proposing fields that do not exist in the client library version being used.
• Troubleshooting and Reporting: The Assistant can diagnose known integration pain points including offline conversion upload failures, manager account hierarchy problems, and Performance Max listing filter issues, then execute validated queries and return results as formatted tables or CSV files.

For operations teams, these features translate into real reliability gains. Bad GAQL frequently surfaces only after a release or during scheduled runs, when it can knock out critical systems. By catching errors earlier in the workflow, the Assistant shifts analytics reliability from a training problem to a tooling problem.

How to Evaluate Google's New Developer Tools for Your Team?

Organizations considering adoption of Google's rebuilt Ads API Assistant should evaluate several factors before rolling the tool across teams:

• Migration Planning: Version 4.0.0 is not compatible with previous releases, so teams need to identify which repositories and scripts will break and plan coordinated upgrades across client libraries, continuous integration scripts, and developer setup documentation.
• Governance and Control: Determine whether the plugin can be installed and updated through existing endpoint management or developer workstation controls, and whether versions can be pinned to prevent silent behavior changes mid-quarter.
• Data Access and Audit: Understand how the Assistant will authenticate to Google Ads accounts for ad hoc reporting and what logging exists for queries executed and CSV exports created.
• Workflow Integration: Decide whether GAQL validation should be required as a pre-commit hook or continuous integration step for teams shipping dashboards, pacing alerts, or conversion pipelines.
• Environment Standardization: If Claude Code or Antigravity becomes the standard host for the Assistant, clarify the support model and determine who owns integration runbooks when troubleshooting spans both AI tooling and Ads API behavior.

Google claims the v4.0.0 architecture cuts token use by pulling in only the rules and schemas required for a given task, and promises quicker replies plus lower CPU and memory use compared to previous versions. However, the larger operational issue is how the plugin fits within enterprise controls and existing developer workflows.

What Does This Reveal About the Future of Enterprise AI?

Both the legal AI and Ads API announcements point to a consistent strategy: Google is betting that enterprise value lies not in models alone, but in reducing friction across entire workflows. By providing validation, integration, and infrastructure alongside models, Google positions itself as a platform that makes AI tools more reliable and easier to deploy at scale.

Weil's partnership with Google on Gemini Enterprise for Legal demonstrates that even the largest, most sophisticated enterprises see value in this integrated approach. Rather than building everything in-house, Weil chose to collaborate with Google on foundational infrastructure, then build specialized applications on top. This pattern suggests that the future of enterprise AI adoption may depend less on which company has the best model, and more on which company can best integrate models into the workflows where lawyers, developers, and analysts actually work.