I’ve watched a lot of corporate maneuvers in my career, but this one caught me off guard.
Anthropic and OpenAI just launched competing joint ventures worth billions. Same week. Same strategy. Same target: the enterprise consulting market that McKinsey and Accenture have owned for decades.
This isn’t about selling software licenses anymore. This is about rebuilding consulting from the ground up, with AI models at the core instead of armies of MBAs with PowerPoint decks.
The Numbers Tell You Everything
Here’s what grabbed my attention. For every dollar companies spend on software, they spend six on services.
That ratio has made consulting a multitrillion-dollar industry. And now two AI labs are positioning to disrupt it with a model that looks nothing like traditional software sales.
Anthropic’s venture pulled in backing from Blackstone, Hellman & Friedman, and Goldman Sachs. The structure: $1.5 billion valuation with $300 million commitments from each founding partner. General Atlantic, Leonard Green, Apollo Global Management, GIC, and Sequoia Capital joined the party.
OpenAI countered hours later with The Development Company, raising $4 billion from 19 investors at a $10 billion valuation. TPG, Brookfield, Advent, and Bain Capital signed on. No overlap between the two ventures.
The logic is identical. Raise capital from alternative asset managers, then get preferred access to their portfolio companies. The investors capture value from resulting contracts while the AI labs get engineering resources to deploy their models at scale.
The Forward-Deployed Engineer Model Goes Mainstream
Remember Palantir? They pioneered the forward-deployed engineer approach by sending technical teams directly into client organizations to build custom solutions.
That model worked because clients got engineers who understood both the technology and their specific business problems. No generic consulting playbook. No cookie-cutter implementation.
Now Anthropic and OpenAI are adopting the same strategy, but with AI models as the foundation.
Goldman Sachs’ Marc Nachmann called it “democratizing access to forward-deployed engineers” for companies that can’t afford the talent or consulting fees to build AI systems independently. Translation? We’re bringing Palantir’s approach to the mass market.
Jon Gray from Blackstone pointed to the real bottleneck, the scarcity of engineers who can implement frontier AI systems at speed. His portfolio companies grew their large language model spending 15-fold over the past year. The demand exists. The supply doesn’t.
These ventures aim to fix that gap.
The Built-In Distribution Advantage
Here’s where it gets interesting from an investor’s perspective.
The partners backing these ventures control access to thousands of portfolio companies. Anthropic’s venture has a pipeline across hundreds of companies through its investor network. OpenAI’s partners have relationships with more than 2,000 portfolio companies and clients.
That’s not a sales pipeline. That’s a built-in distribution network.
Private equity firms pressure their portfolio companies to cut costs and boost productivity. AI deployment pitches map perfectly onto that mandate. The incentives align.
This is similar to how I used to find opportunities—look for companies with natural distribution advantages that competitors can’t easily replicate. These ventures have distribution locked in from day one.
Revenue Models Are Shifting
Both companies are seeing the same trend: enterprise revenue is growing faster than consumer revenue.
OpenAI reported that enterprise now represents more than 40% of revenue and is on track to reach parity with consumer by end of 2026. Anthropic’s enterprise customers represent approximately 80% of revenue, with more than 1,000 businesses spending over $1 million annually.
But here’s what matters: the future of AI revenue might not look like software licensing at all. It might look like consulting, rebuilt from the model up.
Traditional software companies sell licenses and collect recurring subscription fees. These AI ventures are building something different—custom implementations with forward-deployed engineers, ongoing optimization, and deep integration into client workflows.
The margins might be different. The scalability might be different. But if they capture even a fraction of the consulting market, the total addressable market is massive.
The Valuation Arms Race Context
These joint ventures are launching while both companies fundraise at a pace I’ve rarely seen.
OpenAI closed its latest round with $122 billion in committed capital at a post-money valuation of $852 billion. Anthropic received multiple preemptive offers to raise around $50 billion at valuations between $850 billion and $900 billion.
For context: Anthropic’s annualized revenue run rate climbed from roughly $9 billion at year-end 2025 to more than $30 billion by late March 2026. No company in American technology history has grown at that rate.
A potential Anthropic IPO could come as early as October 2026.
These valuations assume continued hypergrowth. The joint ventures provide a path to sustain that growth by opening a new revenue channel that doesn’t depend solely on API calls or subscription seats.
What This Means for the Consulting Industry
McKinsey, Accenture, Deloitte have dominated enterprise consulting for decades. Their model? Send teams of consultants, conduct analysis, deliver recommendations, implement solutions, collect fees.
These AI ventures are building a competing model. Send forward-deployed engineers with frontier AI models, build custom tools integrated into existing workflows, optimize continuously, collect fees.
The traditional firms aren’t standing still. They’re all investing heavily in AI capabilities and partnerships. But they’re also defending existing revenue streams and business models.
The AI labs don’t have that constraint. They can build from scratch with AI-first architectures.
I’ve seen this pattern before—established players with legacy systems and business models getting disrupted by newcomers with different cost structures and approaches. It doesn’t always work. But when it does, the returns can be extraordinary.
The Risks Nobody’s Talking About
Here’s what keeps me cautious about this.
First, implementation risk is real. Building custom AI solutions for enterprise clients is hard. Each deployment is different. Each industry has unique requirements. Scaling that model while maintaining quality is difficult.
Second, the forward-deployed engineer model is expensive. You need talented engineers willing to work on-site with clients. You need them to understand both the technology and the client’s business. That talent is scarce and expensive.
Third, we don’t know the unit economics yet. Traditional consulting has known margins and scalability limits. These AI ventures are experimenting with a hybrid model. Will the margins support the valuations? We’ll find out.
Fourth, client results matter. If early deployments don’t deliver measurable value, word spreads fast in the enterprise market. One bad implementation can poison a dozen potential deals.
What I’m Watching
I want to see three things before drawing conclusions about whether this strategy works.
Case studies and client retention. Are clients renewing engagements? Are they expanding to other parts of their organizations? Client retention tells you if the value proposition is real.
Revenue per engineer. How much revenue can each forward-deployed engineer generate? This metric determines scalability and margins.
Competitive response. How do traditional consulting firms respond? Do they partner with AI labs, build their own models, or try to compete on implementation quality?
The answers to these questions will determine whether we’re watching a fundamental shift in how enterprise AI gets deployed or an expensive experiment that fizzles out.
The Bigger Picture
Step back and look at what’s happening.
Two AI labs are raising billions to build consulting businesses that compete directly with established firms. They’re doing it by offering something different: AI models owned by the vendor, forward-deployed engineers who understand the technology deeply, and access through investors’ portfolio companies.
The traditional consulting model relies on armies of consultants applying frameworks and methodologies. The AI lab model relies on smaller teams of engineers deploying frontier models.
Which approach wins? Probably both, in different contexts.
But if these ventures succeed even partially, they’ll capture billions in revenue that would have gone to traditional firms. And they’ll prove that AI revenue models can look very different from software-as-a-service subscriptions.
That’s worth paying attention to—not because you should rush to invest in either company (both are private), but because it signals where enterprise AI adoption is headed.
The bottleneck isn’t the technology anymore. It’s implementation. Companies that solve the implementation problem at scale will capture enormous value.
I learned a long time ago that the most profitable opportunities often come from solving mundane problems—like helping companies actually use the technology they’ve bought. These joint ventures are betting billions on that thesis.
We’ll see if they’re right.
MarketMagnetix Media Group Insights 
Leave a comment