In the past few years, 75 percent of all the money invested in generative AI (gen AI) has come from the United States and been directed at domestic investments. Given this statistic, some people think Europe has already lost the generative AI battle. We truly believe this is not the case, and that it is not too late for Europe to compete and win in the gen-AI race.

There are three reasons for this belief. First, we have done it before. In 1954, at the dawn of the nuclear energy age, we created the European Organization for Nuclear Research (CERN), where some 6000 scientists laid the foundations of Europe’s world-class nuclear energy industry. That model may apply to AI as well, as I’ll explain.

Second, we already have generative AI champions in Europe like Mistral AI, a French company that produces large language models (LLMs). Mistral was founded by French Silicon Valley engineers who returned home to found the company, and has raised 500 million euros. Similarly, another LLM company in Germany, Aleph Alpha, which was founded by former German employees of other tech firms, has also raised approximately 500 million euros. So it's not too late to leverage these open-source, large language models to compete with American ones.

The third thing Europe has at scale is human capital. We have many colleges, universities, and scientific entities where people are researching and developing AI. We also have a deep talent pool across Europe, which can be leveraged to compete and win in the generative AI race. So we recommend assembling a critical mass of AI researchers from across Europe—just as we did with nuclear scientists at CERN—to create a public-private research center to connect the dots between all the European institutions conducting AI research, ensuring the collaboration necessary to generate innovation and fuel AI start-ups being funded as we speak.