Nvidia CEO Jensen Huang has said he is willing to stop supplying chips to customers who create dangerous artificial intelligence technologies, but he believes the current level of concern about the threat of AI is exaggerated and part of the industry’s "growing pains."
Nvidia CEO Jensen Huang has said he is willing to stop supplying chips to customers who create dangerous artificial intelligence technologies, but he believes the current level of concern about the threat of AI is exaggerated and part of the industry’s "growing pains."
Jensen Huang said this in an interview with Bloomberg during a visit to London.
“This is a hypothetical case, but yes. But before we do that, we need to have a negotiation. We don’t run other people’s companies, and our authority to evaluate how others use our technology is limited. Our technology is universal,” Huang said, answering questions about taking action against irresponsible developers.
Concerns about AI security have led some market leaders, including Anthropic CEO Dario Amodei, to call for a general slowdown in development. However, the Nvidia chief is convinced that the technology can be safely controlled by AI itself, and fears of a market “bubble” are unfounded. According to him, demand for AI infrastructure will exceed supply for several more years.
To help ensure the industry is safe, Nvidia will commit $250 million over five years to UK-based AI security research. The support will come in the form of access to computing power for academic institutions.
