Google’s new AI hub in Paris proves that Google feels insecure about AI

This morning, Google’s CEO Sundar Pichai inaugurated a new hub in Paris dedicated to artificial intelligence. This hub is located in a newly renovated building near Google’s main office in Paris. It will host around 300 researchers and engineers.

But if you pay close attention to Google’s whereabouts in Paris, you may think that Google already launched an AI research center in Paris back in 2018. In fact, the company told me that it doesn’t plan to grow its AI team for this new hub. There’s some new office space, but the team of 300 researchers and engineers who are going to work from the new hub were already working for Google Research and DeepMind — but also YouTube and Chrome.

And yet, several high-profile government members, such as Bruno Le Maire, the minister of the economy, and Valérie Pécresse, the president of the Ile-de-France region, were there to welcome Sundar Pichai and congratulate Google on the announcement.

Given the involvement of Google’s top leader and government members, let’s dissect this communication effort. With the positioning of this announcement, Google wants to remain near the top of the list when it comes to attracting AI talent.

Google could have sent an email to its employees to tell them when they can pick up their badges for the new office. Instead, the company decided that it was a PR opportunity. The company needs to show that it cares about AI and that it’s a priority.

And there’s a reason why Google thinks it can attract AI talent in Paris. Over the past few years, several tech giants have set up AI research labs in Paris. In addition to Google, Facebook (now Meta) created its Paris research lab back in 2015 with Yann LeCun at the helm of the AI initiative — this research group is called FAIR, for Facebook Artificial Intelligence Research.

Since then, many researchers and engineers left their jobs at big tech companies to create startups. Mistral AI is arguably the most emblematic example of that as the young startup has already raised hundreds of millions of dollars to develop new foundational models.

But there’s a thriving ecosystem of AI startups in Paris. Examples include Nabla, Dust, Gladia and Giskard. In just a few years, some of the people working for these startups could also choose to work for Google if they’re looking for something a bit different.

Everything is AI now

At the same time, it shows once again that Google is insecure about artificial intelligence. Given that this building will host teams working not only on AI research projects but also on consumer products like YouTube and Chrome, Google could have simply called it a “Google hub”.

But the company decided that this was an AI hub. They want to say loud and clear that they are an AI company. Sure, the tech giant just launched Gemini Ultra, its most powerful large language model to day. But most people still think about ChatGPT when they think about an AI assistant.

A year ago at a press conference in Paris, Google unveiled Bard, its AI chatbot assistant that is now called Gemini. It was a rushed effort to catch up with ChatGPT.

But it was as much about launching a product as it was about planting a flag that shows that the company is capable of releasing an LLM-based chatbot — and then iterate. You could consider that today’s new AI hub is the continuation of that strategy of regular announcements about AI.

To be fair, Google isn’t the only tech giant making grand gestures about AI investments. In addition to its financial commitment in ChatGPT’s parent company OpenAI, Microsoft also announced today a €3.2 billion ($3.4 billion) investment over the next two years in AI infrastructure in Germany.

Once again, this investment isn’t focused on artificial intelligence exclusively. Microsoft plans to create data centers in Germany for its Azure cloud platform. Some of Azure’s products are focused on AI, but it’s a much larger cloud provider with clients that have nothing to do with the AI industry. So it’s not just a Google thing.