August 18, 2026 · 10:32 Uhr
OpenAI has signed a 20-year contract for an 8-gigawatt data center in Ohio, with Nvidia as the exclusive chip supplier and financing participation of up to $105 billion. The deal marks a paradigm shift: no longer model quality, but computing capacity is the decisive bottleneck in AI competition. For the semiconductor and energy industries, this means massive demand growth for years to come.
Prediction Markets see Anthropic with 98% probability as operator of the strongest AI model by end of August 2026 – a historically high value with $1.6 million trading volume. In parallel, markets forecast an IPO valuation of $1.8 trillion with 72% probability. Anthropic's market dominance over OpenAI and Google has thus established itself as consensus in just a few weeks.
Mistral has secured binding enterprise commitments from ASML, Amadeus, Capgemini and other European heavyweights for a 1-gigawatt compute network, while simultaneously expanding its API infrastructure with regional EU/US endpoints and enterprise SLAs. This is Europe's most ambitious attempt so far to secure technological sovereignty in AI. However, Fortune notes that Mistral's models still lag behind top US and Chinese models.
Crusoe, an AI infrastructure-specialized data center company, is in talks with at least four Wall Street banks – including Goldman Sachs, Morgan Stanley and Bank of America – about an IPO, according to Axios. The move signals that AI infrastructure, after software and chips, is now pushing to market as a standalone asset class. Investors interpret this as confirmation that the infrastructure layer of the AI wave is more sustainably monetizable than model providers.
A current corporate study shows: 93% of surveyed organizations believe that AI agents will deliver measurable ROI by 2028 – yet weak governance, lack of process maturity and insufficient oversight are slowing adoption. Companies are clearly moving away from chatbots toward autonomous agents that act independently. The lack of governance frameworks is becoming the central enterprise risk and regulatory topic for the next 24 months.
AI competition is shifting decisively in August 2026 from the model to the infrastructure level: OpenAI's 8-GW deal with Nvidia and Crusoe's IPO plans show that computing capacity has become a strategic resource – comparable to oil in the 20th century. According to Prediction Markets, Anthropic has established itself by far as the leading model provider, while Mistral is making Europe's last credible attempt to secure AI sovereignty through industrial coalitions. The rapid proliferation of autonomous AI agents in enterprises amid simultaneously lacking governance infrastructure creates systemic risk that is putting regulators worldwide on alert. Geopolitically, the tripartite division of AI space into US, China and European spheres is intensifying, with Europe threatening to fall behind technologically despite Mistral's efforts.
Tokens: 2,156(1,316 in · 840 out)
This website uses cookies. Strictly necessary cookies are always active. By clicking "Accept all" you additionally consent to analytics cookies (Google Analytics). Privacy Policy →
