Most enterprise artificial intelligence projects don’t fail because the technology doesn’t work. They fail because the technology never makes it into the workflow.
That gap between pilot and production is now the defining problem in enterprise AI and the two largest independent AI companies just converged on the same answer. On Monday, OpenAI finalized The Deployment Company, a $10 billion joint venture backed by TPG, Brookfield Asset Management, Bain Capital, Advent International and 15 other investors, Bloomberg reported.
Anthropic announced a parallel enterprise services firm the same week, backed by Blackstone, Hellman & Friedman, Goldman Sachs, General Atlantic, Apollo Global Management and Sequoia Capital.
When Pilots Stall, the Cost of Standing Still Rises
The cost pressure is real and getting harder to ignore. Uber’s chief technology officer, Praveen Neppalli Naga, said in an interview with The Information that the AI budget he’d planned was blown well past projections, driven by surging internal use of Anthropic’s Claude Code. Roughly 11% of live updates to Uber’s back-end systems are now written by AI agents, up from a fraction of a percent three months earlier, according to PYMNTS. Uber’s research and development expenses rose 9% to $3.4 billion in 2025.
That pattern of usage surging past projections while cost controls lag, recurs across enterprise deployments. AI tools don’t carry fixed license fees. Every interaction consumes compute, measured in tokens and organizations that actively push adoption learn that costs scale in kind.
The adoption problem runs deeper than cost management. EY research found that 78% of U.K. organizations believe AI is either fully or mostly implemented across their operations, yet nearly half say their approach is insufficient for the demands of autonomous AI systems, according to EY. Ambition and execution remain misaligned, and the gap is creating urgency at the executive level.
Advertisement: Scroll to Continue
Forward-Deployed Engineer as Sales Force
Both companies’ new ventures share a common operational model: engineers embedded directly inside client organizations. Anthropic’s firm will deploy applied AI engineers from its own staff alongside the new venture’s engineering teams, building custom Claude-powered systems around each company’s existing operations. OpenAI’s Deployment Company will generate revenue through implementation services, product revenue shares and co-owned products across more than 2,000 portfolio companies and clients belonging to its private equity partners.
That model has a name now recognized across the industry: forward-deployed engineering. Monthly job listings for the role grew more than 800% between January and September 2025, according to PYMNTS. In March, Accenture and Microsoft launched a joint forward-deployed engineering practice designed to bring thousands of AI-skilled engineers directly into client environments. EY launched its own forward-deployed engineer roles in the U.K. and Ireland in April. BCG has rebranded engineers within its BCGX unit under the same designation.
Private Equity Becomes the Distribution Channel
The structural logic of both deals centers on private equity’s reach. OpenAI’s partners have access to more than 2,000 portfolio companies across healthcare, logistics, manufacturing and financial services, according to Bloomberg. Anthropic’s venture targets mid-sized companies that lack the in-house resources to build and run frontier AI deployments, including community banks, manufacturers and regional health systems.
The two structures differ in how they compensate investors. OpenAI offers a guaranteed minimum return of 17.5% annually over five years, seniority and downside protection — terms that convert a piece of growth optionality into a fixed-yield-style instrument PE firms can underwrite like a credit fund. Anthropic’s arrangement is common equity with no guaranteed floor, relying instead on the operational credibility of its anchor partners.
Both companies have seen strong demand for AI coding tools and are now pushing into financial services and healthcare to deepen their commercial footprints. OpenAI COO Brad Lightcap has shifted into a new role leading the company’s enterprise push, including overseeing The Deployment Company.
The concurrent announcements signal that the standard enterprise software sales cycle, deal by deal, contract by contract — can’t move fast enough to capture the next wave of AI adoption. Private equity firms, which structurally influence software purchasing across their entire portfolios, offer both companies a faster path from model to workflow.
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
