Arm's CEO said that artificial intelligence might help humanity cure cancer within our lifetime. However, before achieving this goal, the entire industry first needs to address a practical issue: producing enough chips to provide the required computing power for AI.

"I have always believed that the truly disruptive application areas of AI are in healthcare," said Arm CEO Rene Haas in a BBC interview video released on Tuesday. "AI can not only shorten the time required for drug development but also the time needed for drug testing."

"First, we need more wafer fabrication plants"

Haas joined Arm in 2013 and became CEO in February 2022, previously working at NVIDIA for seven years as Vice President of Computing Products. In the interview, he stated that AI could potentially solve some problems that humans alone would find difficult to overcome: "I believe that AI will help humanity cure cancer during our lifetime."

However, he pointed out that this requires massive computing power, and the chips needed for cancer research and humanoid robot development are being consumed in large quantities by AI companies planning to build data centers with gigawatt-level capacity. "Currently, computing power supply is indeed very limited. Before we can build data centers in space, we first need more wafer fabrication plants, and I can tell you that clearly."

Haas said that the demand for chips and memory from AI is at a level the semiconductor industry has never seen before, and building new wafer fabrication plants cannot quickly solve the problem—building one may require hundreds of billions of dollars, and the construction period usually takes two to three years. He said the semiconductor industry is currently in an "absolute environment of supply constraints," and the shortage will continue for some time.

Curing cancer is not just about computing power, and expanding data centers has also sparked opposition

Even if the chip bottleneck is solved, it does not guarantee that AI will cure cancer in the next few years. Medical researchers point out that cancer is not a single disease but a large group of more than 100 diseases, and there are significant differences between patients' cancers. AI can indeed help find potential drug targets and predict drug effects, but evidence of its actual impact on clinical medicine remains limited, and the vast number of variables within the human body make it difficult for models to reliably predict the effectiveness of candidate drugs.

Other chip executives have also issued similar warnings. Manish Bhatia, Chief Operating Officer of Micron, said in January this year that the current memory shortage "is indeed unprecedented," with HBM consuming a large amount of capacity, causing "significant supply shortages" for traditional products such as phones and PCs; Qualcomm CEO Cristiano Amon said in a February earnings call that "the memory shortage and price increases across the industry" will determine the scale of the smartphone market this year.

In recent years, the expansion of data center construction in the United States has been rapid, with over 1,400 data centers completed or approved for construction by the end of last year. As data centers continue to enter communities, their impact on air quality and water resources has sparked controversy. A Gallup survey in May this year showed that 70% of 1,000 American adults oppose building data centers for AI in their local area, with nearly half expressing "strong opposition."