Voice of AI - Clarity instead of hype: When even OpenAI admits that the model isn't enough 🧭

There are moments when the industry announces its own turning point without explicitly labeling it as such. This was one of them. On May 4, 2026, OpenAI and Anthropic founded two joint ventures with the world's largest private equity firms, within hours of each other . Their sole purpose: to finally bring artificial intelligence to productive use in businesses. When the very inventors of the models admit that the model alone creates no value, the hype cycle is officially over. The phase of delivering results begins.

The multi-billion dollar admission

OpenAI raised four billion dollars from TPG, Brookfield, Advent, and Bain Capital for a vehicle aptly named "The Deployment Company," valued at ten billion dollars. Simultaneously, Anthropic partnered with Goldman Sachs, Blackstone, Hellman & Friedman, Apollo, General Atlantic, and GIC to create a 1.5 billion dollar structure that directly integrates Claude into his partners' portfolio companies. Both structures deliberately replicate Palantir's forward-deployed engineer model and target the same 375 billion dollar market currently dominated by McKinsey, BCG, and Accenture.

Nicholas Lin, Head of Product for Financial Services at Anthropic, formulated the real admission with unusual clarity: there is a huge gap between what AI can do today and the value the market actually derives from it. Marc Nachmann of Goldman Sachs added precisely that the problem isn't a lack of models, but a lack of people who can translate these tools into business processes. This is exactly where OAK AI has been operating since day one. Clarity instead of hype isn't just a pose, but the only viable methodology at this stage.

The M&A wave is gaining momentum

CB Insights counts 266 completed AI acquisitions in the first quarter of 2026, an increase of approximately 90 percent compared to the same quarter of the previous year. IREN signed the acquisition agreement for Mirantis on May 5th for $625 million in stock, thereby securing Kubernetes expertise and 1,500 enterprise customers. Nebius acquired Eigen AI to optimize inference—precisely the area where AI models in production both cost and generate revenue. SoundHound AI is pursuing the end-to-end conversational platform with its announced acquisition of LivePerson.

The message is unambiguous. Those who don't build their own facilities by 2026 will be bought out. Those who don't buy will fall behind. This applies to hyperscalers as well as to European SMEs, which must strategically defend their sovereignty right now.

Nvidia as a center of gravity

CNBC reported on May 11 that Nvidia has already committed over $40 billion in AI equity investments in the first months of 2026. Of that, $30 billion is earmarked for OpenAI alone, with billions more flowing into companies including Corning and data center operator IREN. Critics point to this circular logic. In reality, we are witnessing the creation of a virtually unassailable ecosystem that will define which supply chains and partnership models will remain viable over the next three years. Anyone strategically sourcing AI today must be able to understand these interconnections before making a decision.

Geopolitics becomes a sourcing issue

At the end of April, Beijing's National Development and Reform Commission blocked Meta's planned $2 billion acquisition of the AI agent company Manus. This ban breaks the so-called Singapore-washing route, which founders had used to evade oversight in both Beijing and Washington. For European investors and buyers, this is a wake-up call. AI sourcing in 2026 no longer simply means choosing the best model. It means assessing regulatory requirements, data sovereignty, and supply chain risks based on facts. Tech due diligence is transforming from a technical add-on to a core strategic competency.

The 95 percent gap is the real market.

Gartner forecasts global AI spending of $2.52 trillion for 2026, a 44 percent increase year-over-year. Generative models are growing at 80.8 percent, and AI-optimized servers at 49 percent. At the same time, a widely cited MIT study shows that 95 percent of companies have not yet achieved a measurable return on investment in generative AI. This discrepancy is not the problem; it's the opportunity. Those who take a structured approach now will secure advantages that money will no longer be able to buy in two years.

What the Frontier laboratories are now openly saying

Even Anthropic, which empowers agent systems for continuous self-improvement with its new "Dreaming" technology, knows that self-optimizing agents scale not only value but also errors. They need governance, clean data architecture, and an honest assessment of their current state before going into production. This is precisely when independent evaluation proves its worth.

🎯 Key takeaways of the week

• OpenAI and Anthropic, with $10 billion and $1.5 billion respectively, admit that deployment is the real bottleneck, not the model.

• 266 AI M&A deals in Q1 2026, an increase of 90 percent, signal acute strategic pressure for action, also for medium-sized businesses.

• Nvidia is investing over $40 billion in AI equity, thereby defining the sourcing landscape for the next three years.

• 95 percent of companies see no measurable AI ROI; the market for structured implementation is exploding.

• Geopolitics and data sovereignty make independent tech due diligence a mandatory discipline.

Sources

• TechCrunch, "Anthropic and OpenAI are both launching joint ventures for enterprise AI services", May 4, 2026 → https://techcrunch.com/2026/05/04/anthropic-and-openai-are-both-launching-joint-ventures-for-enterprise-ai-services/

• CNBC, "Anthropic teams with Goldman, Blackstone and others on $1.5 billion AI venture", May 4, 2026 → https://www.cnbc.com/2026/05/04/anthropic-goldman-blackstone-ai-venture.html

• The AI Insider, "Nvidia Commits Over $40B to AI Equity Deals in 2026", May 11, 2026 → https://theaiinsider.tech/2026/05/11/nvidia-commits-over-40b-to-ai-equity-deals-in-2026-led-by-30b-openai-investment/

• Yahoo Finance, "IREN Buys Mirantis to Boost Infrastructure Capabilities", May 5, 2026 → https://finance.yahoo.com/sectors/technology/articles/ai-cloud-acquisition-expands-iren-142551901.html

• FE International / CB Insights, "AI M&A Trends 2026: Why Acquirers Pay Premium Multiples", April 2026 → https://www.feinternational.com/blog/ai-ma-trend

• Technology.org , "China Hits Unfriend on Meta's $2 Billion Manus AI Acquisition Deal", April 28, 2026 → https://www.technology.org/2026/04/28/china-hits-unfriend-on-metas-2-billion-manus-ai-acquisition-deal/

• Axios, "Why private equity is making deals with the AI giants", May 5, 2026 → https://www.axios.com/2026/05/05/openai-anthropic-private-equity

Conclusion

The Frontier labs have done this week what has been overdue in European SMEs for months: they've stopped celebrating the model and started organizing deployment. OAK AI has provided this clarity since day one.

Our AI Readiness Check provides a structured assessment of your current position with peer benchmarking, clear priorities, and a realistic roadmap. Our AI Sourcing approach evaluates technologies, providers, and investments based on facts, independently, and with decision-making relevance. Our platform demonstrates how AI can be transformed from hype into measurable value.

👉 Schedule a live demo of the AI Impact platform: https://www.oakai.de/ 👉 Request an AI Readiness Check: info@oakai.de

Those who wait until 2026 will be bought out in 2028.

The future is not a matter of chance. It is a choice.

Ralph Schwehr CEO & Chief AI Officer, OAK AI

Here are current screenshots of the results from our tools...