AI Bubble Risk: Hassabis on Google’s Resilience

The Financial Times (FT) reported on the 24th (local time) that Demis Hassabis, CEO of Google DeepMind and winner of the 2024 Nobel Prize in Chemistry, warned that excessive enthusiasm in some parts of the artificial intelligence (AI) industry is increasingly looking like a bubble.

In a recent interview with the FT in Davos, Switzerland, CEO Hassabis said, “It seems somewhat unsustainable that a new startup with no product, no technology, or anything yet, is receiving billions of dollars in seed investment. This trend may lead to a correction in some parts of the market.”

The FT explained that Hassabis’s remarks are significantly different from the recent dismissal of concerns about excessive investment in AI by other technology industry leaders, such as NVIDIA CEO Jensen Huang and Microsoft CEO Satya Nadella, in Davos.

For example, the venture capital industry is rushing to invest in companies such as Thinking Machine Lab, founded by Mira Murati, a former Open AI executive. The company was valued at $10 billion just six months after its founding, but has revealed little about what it is developing. The company has recently lost some of its key personnel, raising questions about its long-term business sustainability.

Hassabis emphasized that demand for AI is stronger than ever across the company’s products, including Google’s latest Gemini 3 model, and said AI is “probably the most transformative technology ever invented.”

He emphasized, “Even if the bubble bursts, we will be fine. We already have a great business foundation that can increase productivity by adding AI capabilities.”

Google is evaluated to have succeeded in rebounding last year after overcoming the difficult times it experienced after the launch of Open AI’s ChatGPT in 2022. Currently, Google’s AI model performance is surpassing that of smaller competitors and narrowing the gap in the number of chatbot users.

Thanks to this trend, the market capitalization of parent company Alphabet recently exceeded $4 trillion, making it the second most valuable company in the world after semiconductor company Nvidia.

Hassabis also argued that Western companies are still ahead of China in the race to develop AI.

China’s DeepSearch surprised Silicon Valley about a year ago by developing a powerful, free-to-use AI model at a much lower cost than its U.S. competitors. The emergence of this model also shocked the U.S. stock market, where capital is concentrated in big tech stocks.

Hassabis said, “The West has overreacted,” and added, “Chinese research labs have not yet proven their ability to innovate beyond the frontier. American technology companies maintain a technological advantage of about six months.”

However, over the past year, China has made large investments in developing leading AI models that developers can freely use, and is evaluated to have surpassed American competitors in the open AI field.

“Chinese companies are more focused on short-term applications and immediate revenue generation,” Hassabis said. “The research-driven, front-line capabilities needed to achieve artificial general intelligence (AGI) beyond human capabilities are still the goal of US-based companies—DeepMind, OpenAI, Anthropic.”

At the Davos Forum, the risks and side effects surrounding AI were also major topics of debate. Hassabis said it is essential to focus on developing safe and responsible AI, and that the AI ​​industry must clearly demonstrate the positive value of AI to the general public.

“For us, it’s about doubling down on AI for science and AI for medicine, and those are clearly good things for the world,” he said.

Hassabis noted that Google’s AI model could help realize the company’s long-held vision of smart glasses. Google first introduced smart glasses more than 10 years ago, but at the time, the devices were widely ridiculed and ignored by consumers.

Last year, Google announced that it would collaborate with fashion companies such as Warby Parker to introduce new smart glasses incorporating AI.

“I think we were a little ahead of our time when we first started this device at Google about 10 years ago,” Hassabis said. “At the time, there was no definitive killer app. I think a universal digital assistant that helps us in our daily lives could be that killer app.”

Hassabis has greatly expanded his authority and responsibility for Google’s overall AI business in recent years, and is evaluated as a key figure in Google’s future strategy.

However, he drew a line at the possibility of succeeding Sundar Pichai, who also serves as CEO of Google and Google’s parent company Alphabet, in the future, saying “no.” “I am very satisfied with what I am doing now,” he explained. “I like being close to science and research. There is a limit to what you can do in a day, and you also need time to think deeply.”