Anthropic has formed a $1.5bn joint venture with private-equity and bank partners to sell tailored AI services to companies. The deal names Blackstone, Hellman & Friedman and Goldman Sachs among founding partners and includes a string of other investors. Hours earlier Bloomberg reported OpenAI was raising $4bn for a similar vehicle called The Development Company at a $10bn valuation. Both moves create new channels for selling AI into corporate clients and put capital and engineers closer to end users.

Anthropic announced the joint venture on Monday. TechCrunch reported the new company was valued at $1.5bn, and that Anthropic, Blackstone and Hellman & Friedman each committed $300m. The announcement listed Goldman Sachs as a founding partner and named backers including Apollo Global Management, General Atlantic, GIC, Leonard Green and Sequoia Capital.

Anthropic described the venture as a way to work directly with customers. "An engagement might begin with the company’s engineering team sitting down with clinicians and IT staff to build tools that fit into the workflows that staff already use," the company said in its announcement. "Engagements like this will run across mid-sized companies across industries, each shaped by the people closest to the work."

OpenAI moves on a larger scale

Bloomberg reported that OpenAI is raising money for a sibling vehicle called The Development Company. That effort would raise about $4bn from 19 investors against a $10bn valuation, the report said. Named investors include TPG, Brookfield Asset Management, Advent and Bain Capital.

The two funds have different sizes. Anthropic's is smaller by headline numbers. OpenAI's is larger and involves a broader investor pool. The two ventures appear to run on the same idea: marry deep-pocketed alternative asset managers with the AI labs that build the models.

Why investors are joining

Both announcements and reporting on them frame the ventures as channels to sell enterprise services into investors' portfolio companies. The ventures are structured so investors can get early or preferred access to deals inside the firms they own or advise.

That gives private-equity firms and hedge funds a way to capture more value from AI contracts.

Mastheads

The new capital will also pay for engineering teams devoted to individual clients. TechCrunch and other coverage highlight the forward-deployed engineer model, where engineers embed with a customer to build bespoke tools. The approach mirrors practices used by some enterprise software firms, putting engineers on the ground with clients rather than selling off-the-shelf licences alone.

Both labs have been raising money rapidly this year. OpenAI announced $122bn in new funding at the end of March, against a valuation of $852bn, the reporting said. TechCrunch reported that Anthropic is in late-stage talks on a round that would add $50bn of funding at a $900bn valuation. Those headline numbers show how much capital is circling the biggest model builders.

Bringing private-equity partners into enterprise sales changes incentives. Investors gain routes to deploy AI into their portfolio firms.

AI labs gain customers and deep pockets for customised deployments. The arrangement also aligns enterprise work with investor timelines for improving portfolio-company performance.

That alignment matters. Private-equity firms focus on operational change in the companies they own. They often look for software and services that can lift margins or speed growth. A direct line to a company building generative AI tools makes those projects quicker to start and, in theory, easier to measure.

How the deals differ in reach

The reported scale gap matters for execution. A $4bn vehicle gives OpenAI scope to hire larger teams and to underwrite bigger multi-company programmes. A $1.5bn vehicle is smaller but still large enough to fund many bespoke deployments across mid-sized firms.

The investor lists barely overlap. That reduces the chance of the same portfolio company getting preferred offers from both ventures. It also means each lab can court different investor networks and sectors.

Both labs will need to staff and operate these new units. That requires engineers who can translate model capabilities into workflows. It also needs sales and project-management teams that can work inside regulated industries such as healthcare and finance.

Anthropic's announcement stressed that work with clinicians, IT staff and other operational teams. That language signals the venture is targeting regulated sectors that value integration and safety over raw model performance.

Related Articles

TechCrunch reported Anthropic, Blackstone and Hellman & Friedman each committed $300m to the new venture, which was valued at $1.5bn.

This article was created with AI assistance.