All Posts

Successful Hyperscaler IPO: SWI Capital has listed on Euronext Amsterdam at a €1.6B valuation and stock rising 14% in 2 days. European capital markets are finally catching on to the Hyperscale trend.

AI is not yet trusted to run operations forecasting. At a panel on AI in CPG operations, leaders agreed they haven’t encountered an AI model they’d rely on for core forecasts, despite growing complexity from global supply chains, tariffs, and regulation. AI is used tactically, but when failure can halt the business, low-variance, trusted models still win.

In traditional startups, early departures are often survivable because value accrues to product, customers, or revenue. In unicorn-at-birth AI startups, valuation is frequently anchored to people. When those people leave — especially co-founders — the asset itself is diminished.

Europe’s data center construction market is projected to grow from $16.3B in 2024 to $27.9B by 2033, driven by cloud adoption, AI workloads, and sustainability mandates. As highlighted in recent TheMarketAI coverage of SWI Group and Nebius, Europe’s buildout is increasingly defined by energy efficiency, sovereign infrastructure, and long-term resilience rather than pure scale.

A new “unicorn at birth” has entered the AI race. Unconventional AI, a two-month-old startup founded by former Databricks AI chief Naveen Rao, has raised $475M at a $4.5B valuation in a seed round. Backed by a16z, Lightspeed, Lux, DCVC, and Jeff Bezos, the company is targeting one of AI’s hardest problems: building a more energy-efficient computer for next-generation AI workloads as power and compute constraints intensify.

Cisco has invested in World Labs, the spatial-intelligence startup founded by Dr. Fei-Fei Li, underscoring a major shift in AI: from language-based models to systems that understand and operate within 3D environments. World Labs’ Large World Models aim to move AI from “understanding words to understanding worlds.” As spatial and multimodal intelligence emerges, it may define AI’s next decade far more than text-only LLMs.