Wake-up call for the West: Google legend Schmidt warns Europe about the AI ​​revolution

“According to the former Google CEO, only one project should be important for Europe: becoming independent of foreign AI models with the help of open source. This requires more than just billions in investment.

 

Former Google CEO Eric Schmidt traveled to the World Economic Forum in Davos with a wake-up call for Europe. Europe must invest many billions of euros very quickly to create an artificial intelligence (AI) model based on open-source software. This requires a great deal of energy, which Europe currently lacks, and a great deal of hardware, which must be purchased, among other places, from the United States.”

 

Is this practical without cheap and stable Russian energy supply?

 

Schmidt’s proposal for European AI independence is considered challenging but theoretically possible, as Europe shifts its focus from cheap Russian fossil fuels to a much more expensive and sovereign energy strategy as of 2026.

The practicality of this plan without Russian energy relies on several critical shifts in 2026:

1. Energy Costs and Training Hubs

 

    High Costs: Energy in Europe remains significantly more expensive than in the U.S. (twice as high in some cases), leading many European AI projects to train models abroad.

 

    Strategic Relocation: To find "cheap and stable" power, AI training is moving toward secondary markets with abundant renewables, such as Norway (hydroelectric) and Poland, rather than traditional hubs like Frankfurt or Amsterdam.

 

2. Diversification from Russian Energy

 

    Permanent Shift: As of 2026, the EU has largely internalized the loss of Russian energy as a permanent structural change.

    Supply Reduction: Russian gas imports dropped from 45% in 2021 to 19% by late 2024, with the EU aiming for a total phase-out by 2027.

 

    Alternative Stability: Stability is being sought through increased renewable targets (45% of energy mix by 2030) and new partnerships with the U.S. and Gulf states, though these often lack the low price point of former Russian pipeline gas.

 

3. Grid and Infrastructure Bottlenecks

 

    Grid Constraints: The primary obstacle in 2026 is not just the source of energy but the infrastructure. Grid connection delays for data centers are a major hurdle, with vacancy rates for data center capacity hitting all-time lows of 6.5% due to these bottlenecks.

    Legislative Action: The EU's proposed Cloud and AI Development Act and the Data Centre Energy Efficiency Package (due Q1 2026) are intended to triple processing capacity while mandating carbon-neutral operations by 2030 to manage these energy demands.

 

4. Hardware Independence

 

    U.S. Reliance: Schmidt notes that even with energy independence, Europe remains reliant on U.S.-designed hardware (chips), which requires massive capital investment that the EU is currently trying to secure through public-private partnerships.

 

Schmidt’s proposal for European AI independence is considered theoretically possible but unrealistic.