hugging face ceo says we’re in an ‘llm bubble,’ not an ai bubble

The Current State of AI: An "LLM Bubble"?

Clem Delangue, co-founder and CEO of Hugging Face, proposes that the current excitement isn't an AI bubble overall, but rather a bubble specifically surrounding Large Language Models (LLMs). He addressed this at an Axios event on Tuesday, acknowledging the widespread debate about potential market corrections.

LLMs Receiving Disproportionate Attention

Delangue believes that while an LLM bubble may be nearing its peak, the future of Artificial Intelligence remains secure. He suggests that LLMs – the technology behind platforms like ChatGPT and Gemini – are currently attracting an excessive amount of focus.

“I anticipate the LLM bubble could burst within the next year,” Delangue explained. “However, it’s crucial to remember that LLMs represent only a segment of the broader AI landscape, encompassing applications in fields like biology, chemistry, image processing, and audio/video analysis.”

The Rise of Specialized Models

A key argument presented by Delangue is that LLMs aren't universally applicable. He foresees a shift towards smaller, more specialized AI models gaining prominence in the future.

“Currently, there’s a concentration of resources on the idea of a single, massively powerful model solving all problems,” Delangue stated. “The reality will likely involve a proliferation of models tailored to specific tasks and industries.”

Practical Applications and Efficiency

To illustrate his point, Delangue used the example of a chatbot for a banking institution.

“Such a chatbot doesn’t require philosophical capabilities; a smaller, specialized model would be more cost-effective, faster, and potentially deployable on existing enterprise infrastructure,” he clarified. “This represents the future direction of AI development.”

Hugging Face's Position and Financial Prudence

Delangue conceded that a potential LLM bubble burst could have some impact on Hugging Face. However, he emphasized the diversification within the AI industry as a mitigating factor.

Furthermore, Hugging Face maintains a substantial portion – approximately $200 million – of the $400 million in funding it has secured. This conservative financial strategy distinguishes it from many other companies operating within the LLM space.

A Long-Term Vision

“In the context of AI, retaining funds is akin to achieving profitability, given the billions of dollars being invested by competitors,” Delangue noted.

Hugging Face is adopting a more capital-efficient approach, avoiding the rush and potential panic observed in other areas of the AI sector.

“Having 15 years of experience in AI, I’ve witnessed these cycles firsthand,” Delangue added. “We are applying those lessons to build a sustainable and impactful company for the long term.”