Anthropic and OpenAI Both Launch Enterprise AI Joint Ventures With Wall Street Asset Managers
Anthropic launched a $1.5B JV with Goldman Sachs, Blackstone, and Hellman & Friedman to embed AI inside PE-backed companies. OpenAI announced The Development Company, a $10B JV raising $4B from 19 investors including TPG, Brookfield, and Bain Capital.
Anthropic and OpenAI simultaneously announced enterprise joint ventures on Monday — separate structures, same strategic logic. Both AI labs are partnering with major asset managers to create dedicated channels for selling AI services to portfolio companies, with ventures capturing preferred distribution access in exchange for investor capital. The moves signal a major shift in how frontier AI labs plan to monetize at enterprise scale.
Two Ventures, One Playbook
Anthropic's venture, valued at $1.5 billion, was formed with Blackstone, Hellman & Friedman, and Goldman Sachs as founding partners, joined by Apollo Global Management, General Atlantic, GIC, Leonard Green, and Sequoia Capital. Each founding partner committed $300 million. Rather than acting as a traditional AI consultancy, Anthropic's venture will embed engineers inside portfolio companies to redesign core workflows and integrate Claude models at the process level — a hands-on deployment model designed for enterprises that cannot easily buy off-the-shelf AI solutions.
OpenAI's venture, named The Development Company, is operating at a larger scale: the company is raising $4 billion from 19 investors against a $10 billion valuation. Named participants include TPG, Brookfield Asset Management, Advent, and Bain Capital. The Development Company will deploy OpenAI's models — including GPT-4o and the o-series reasoning models — within portfolio company workflows, with a focus on revenue-generating applications and operational transformation.
The Numbers
The capital raised by both ventures is not dilutive equity in the AI labs themselves; it is working capital for the joint venture entities. The logic is asymmetric: investors get preferential access to AI deployment inside their portfolio companies, plus a share of the revenue generated from those deployments. The AI labs get non-dilutive capital and guaranteed distribution through the private equity channel — roughly 25,000 portfolio companies globally hold an estimated $10 trillion in enterprise value, most of which has never been the target of a serious AI deployment effort.
The simultaneous announcements suggest both labs reached the same strategic conclusion at roughly the same time: distribution, not model capability, is the binding constraint on enterprise AI monetization. Private equity firms own more enterprise customer relationships than any other financial intermediary, and they have strong incentives to increase the operating efficiency — and therefore exit valuations — of their portfolio companies.
TechCrunchOur Take
The private equity channel is a smart bet for both labs, but it comes with real risks. PE portfolio companies are acquired for operational leverage, not for greenfield AI transformation — they often have legacy tech stacks, underfunded IT, and management teams focused on EBITDA, not AI roadmaps. The embedded engineer model Anthropic is using is expensive and hard to scale; OpenAI's more arm's-length approach is easier to scale but harder to differentiate. The venture that wins will be the one that builds repeatable playbooks for specific industries — healthcare-revenue-cycle, industrial-supply-chain, financial-reporting — rather than general AI transformation. Watch which industry verticals each venture announces wins in first.
Want frontier-model launches and AI-agent news first? Subscribe to the daily brief →
