ARM CEO Hasse said in an interview with the BBC that AI has transformative potential in healthcare, capable of shortening drug development and testing times, and may help humanity conquer cancer within this generation.

Author and source: BBC

The CEO of Arm said that AI might help humanity conquer cancer within our lifetime. However, before this goal can be achieved, the entire industry must first address a practical challenge: producing enough chips to provide the computing power required by AI.

"I've always believed that the most transformative application of AI is in healthcare," said Rene Haas, CEO of Arm, in a BBC interview video released on Tuesday local time. "AI can not only shorten the time required for drug development but also reduce the time needed for drug testing."

First, more fabs are needed.

Hass joined Arm in 2013 and became CEO in February 2022, after spending seven years at NVIDIA as Vice President of Computing Products. In an interview, he stated that AI has the potential to solve problems that are difficult for humans to overcome on their own: “I believe that within our lifetime, AI will help humanity conquer cancer.”

But he pointed out that this requires massive computing power, and the chips currently needed for cancer research and humanoid robotics development are being heavily consumed by AI companies planning to build data centers with several gigawatts of capacity. “Currently, computing supply is indeed extremely limited. Before we can build data centers in space, we first need more fabs—I can tell you that definitively.”

Hass stated that the scale of demand for chips and memory from AI is at a level the semiconductor industry has never before encountered, and building new fabs cannot quickly resolve the issue—each fab may require hundreds of billions of dollars in investment and typically takes two to three years to construct. He described the semiconductor industry as currently operating in an environment of absolute supply constraints, with tight supply expected to persist for some time.

Curing cancer isn't just about computing power, and the expansion of data centers has also sparked opposition.

Even if the chip bottleneck is resolved, it does not guarantee that AI will cure cancer in the coming years. Medical researchers point out that cancer is not a single disease but a vast group of over 100 distinct diseases, with significant variations between patients. While AI can indeed help identify potential drug targets and predict drug effects, evidence demonstrating its real-world impact on clinical care remains limited, and the vast number of variables within the human body makes it difficult for models to reliably predict the actual effectiveness of candidate drugs.

Other chip executives have issued similar warnings. In January, Micron’s Chief Operating Officer, Manish Bhatia, stated that the current memory shortage is “truly unprecedented,” with HBM consuming significant capacity and causing “severe supply shortages” for traditional products such as smartphones and PCs; in February, Qualcomm CEO Cristiano Amon said on the earnings call that “industry-wide memory shortages and price increases” will determine the scale of this year’s smartphone market.

In recent years, the construction of data centers in the United States has expanded rapidly, with over 1,400 data centers either 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 become a point of contention. A Gallup survey conducted in May this year found that 70% of 1,000 U.S. adults oppose the construction of AI-powered data centers in their local areas, with nearly half stating they “strongly oppose” such projects.

","author":"ME Selection","type":"Repost"}