Anthropic and Blackstone Bet Big on the Business of Making AI Actually Work

AI models keep getting smarter, but the big question is how companies will actually use them. Now, Anthropic and OpenAI are both betting that the real money isn't in building better models—it's in sending engineers into corporate offices to help clients figure out what to do with them. That bet just got a name: Ode.

Ode is a $1.5 billion joint venture launched in May by Anthropic, Blackstone, Hellman & Friedman, and Goldman Sachs. The idea came from Blackstone, which struggled to get its portfolio companies to adopt AI using big consulting firms and small boutiques. One of those boutiques, Fractional AI, impressed the investors so much that the joint venture bought it. Fractional had been working with OpenAI until the acquisition, but now its team forms the core of Ode—a kind of elite, scaled-up AI services firm.

Ode's CEO, Chris Taylor, says it's easy to imagine the company becoming a trillion-dollar business someday. The challenge, he adds, is growing fast without losing quality. Ode currently has 100 engineers, many of them former startup founders, and works closely with Anthropic's applied AI team. The firm uses Anthropic's Claude technology by default, but will switch to rival AI products if needed. Ode's chief technologist, Eddie Siegel, insists the real value isn't in picking the right model—it's in building custom systems that solve specific business problems. “Model selection matters, but it's not where the majority of calories are spent,” he said.

The venture faces stiff competition from OpenAI's own deployment arm and from giants like Deloitte and Accenture. But Taylor believes non-AI companies will be the big winners if they adopt the technology correctly—and that requires top talent most companies don't have. Ode's engineers are described as “grown-up” generalists, the special forces of applied AI. Whether enough of them exist to meet demand is an open question, but Siegel is optimistic. “It has never been an easier time to become an entrepreneur,” he said, noting that the kind of end-to-end problem-solving that founders learn is exactly what Ode needs. If he's right, the next great AI race won't be about the models themselves, but about who can make them work inside the world's largest companies.