TL;DR
OpenAI and Anthropic made separate May 2026 moves into enterprise AI services, pairing frontier models with embedded engineering teams. The confirmed launches show both labs trying to solve the same adoption problem: customers can buy AI tools, but many still struggle to put them into production.
OpenAI and Anthropic have moved from selling AI model access toward building enterprise deployment businesses, with Anthropic announcing a new AI services firm on May 4 and OpenAI launching the OpenAI Deployment Company on May 11, a pair of moves that could reshape how companies buy and implement generative AI.
Anthropic, Blackstone, Hellman & Friedman and Goldman Sachs announced a new AI-native enterprise services firm designed to bring Claude into companies’ core operations. The source material describes the venture as backed by about $1.5 billion and aimed at mid-market businesses, including portfolio companies tied to the investor group.
OpenAI said its Deployment Company will be majority-owned and controlled by OpenAI, backed by more than $4 billion in initial investment and built with 19 investment, consulting and systems-integration partners. The company also agreed to acquire Tomoro, an applied AI consulting and engineering firm, bringing about 150 forward-deployed engineers and deployment specialists into DeployCo from the start.
Axios reported that DeployCo is launching at a $10 billion pre-money valuation. OpenAI’s own announcement does not list that valuation, so it should be treated as reported financial detail rather than company-disclosed terms.
Why It Matters
The moves matter because they show leading AI labs moving into the services layer, where enterprise AI projects often stall. Selling model access is only one part of adoption; companies also need security reviews, data connections, workflow redesign, evaluation systems and staff training before AI can affect day-to-day operations.
The strategy also changes the competitive map. The labs are not only competing with each other on model quality; they are moving closer to systems integrators, consultants and software vendors that already manage enterprise change. If the model works, labs could turn deployment work into recurring model usage and deeper customer lock-in. If it fails, they may inherit the lower-margin labor burden of consulting.
Your AI Survival Guide: Scraped Knees, Bruised Elbows, and Lessons Learned from Real-World AI Deployments
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
The source material frames both launches as a shift toward the Palantir-style forward-deployed engineer model, in which engineers work inside customer organizations, study workflows and build production systems around a platform. That framing is analysis, not a standalone disclosure from both companies.
The business case rests on a widely cited services gap: companies spend far more on implementation and process change than on software licenses alone. MIT’s 2025 GenAI Divide research has also been cited across the industry for finding that most generative AI pilots fail to produce measurable business returns, with poor integration listed as a major reason.
“It will launch with more than $4 billion of initial investment”
— OpenAI, May 11 announcement
“rapidly bring Claude into their core business operations”
— Anthropic partnership announcement
“The challenge now is helping companies integrate these systems into the infrastructure and workflows that power their businesses.”
— Denise Dresser, OpenAI chief revenue officer
“the model isn’t the bottleneck, deployment is”
— Thorsten Meyer AI source analysis
Agentic Architectural Patterns for Building Multi-Agent Systems: Proven design patterns and practices for GenAI, agents, RAG, LLMOps, and enterprise-scale AI systems
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
Several details remain unclear. OpenAI has not disclosed all financial terms behind DeployCo in its own announcement, and the Tomoro deal still needs to close. It is also not yet clear whether these embedded engineering teams will scale like software, with repeatable deployment patterns and stronger margins, or behave more like consulting, where each customer requires a large amount of custom labor.
The timing also needs precision: the source material describes the two moves as occurring within 72 hours, but the public announcements reviewed here show Anthropic’s announcement on May 4, 2026, and OpenAI’s on May 11, 2026.
Workflow Automation with Microsoft Power Automate: Design and scale AI-powered cloud and desktop workflows using low-code automation
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
The next milestones are the closing of OpenAI’s Tomoro acquisition, early customer deployments by both new entities and evidence on whether the forward-deployed model can produce repeatable products rather than one-off implementation projects. Investors and enterprise buyers will also be watching whether other private-equity firms, consultancies and AI labs join similar structures.
AI Programming Made Practical: A Step-by-Step Guide to Building AI-Powered Applications, Writing Better Code Faster, and Using Modern AI Tools with Confidence
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What exactly did OpenAI announce?
OpenAI announced the OpenAI Deployment Company, a majority-controlled entity backed by more than $4 billion in initial investment. It also agreed to acquire Tomoro, adding about 150 forward-deployed engineers and deployment specialists after closing.
What did Anthropic announce?
Anthropic announced a new AI-native enterprise services firm with Blackstone, Hellman & Friedman and Goldman Sachs. The firm is intended to help companies deploy Claude inside core business operations.
Why are AI labs moving into services?
The confirmed announcements point to the same problem: many companies can access powerful models, but cannot easily connect them to data, controls, workflows and production systems. Services work is where much of that adoption happens.
Is this the same as consulting?
Partly. The forward-deployed engineer model resembles consulting because it places technical staff close to the customer. The labs’ bet is that the work will create reusable software patterns, recurring model usage and deeper customer relationships, rather than remaining mostly custom labor.
What remains unproven?
The open question is scale. If deployment teams can turn repeated customer problems into standardized products, the model could support strong enterprise revenue. If every deployment needs proportional engineering time, margins may look more like consulting than software.
Source: Thorsten Meyer AI
