IT Home On October 18, Jack Clark, co-founder of Anthropic, said on Saturday local time that rapid progress in artificial intelligence capabilities will drive more rapid changes across various industries. The company’s internal model predicts that under certain conditions, by 2030, AI could increase U.S. economic output by nearly 50%.
Clark said during a speech at the International Monetary Fund and World Bank annual meeting in Bangkok that international cooperation is needed to establish “unusual and strict rules” for this transformative technology. The window for policy makers to act is gradually narrowing.
Clark said that the estimated U.S. GDP will reach 44 trillion US dollars (IT Home note: at current exchange rates, this is about 294.93 trillion RMB), while the current GDP is slightly above 30 trillion US dollars (at current exchange rates, this is about 201.09 trillion RMB). “I know this number sounds extremely absurd,” but he added that AI is greatly accelerating scientific discoveries.
He mentioned that the premise of this prediction is that “technology continues to improve and evolve. If the GDP achieves such a significant growth, economic theory suggests that new industries and inventions must emerge in order to support such economic expansion.”
When Clark made these remarks, Anthropic was preparing for its initial public offering expected in November, with a valuation of up to $2 trillion (equivalent to approximately 13.41 trillion yuan at current exchange rates).
Currently, humans still have control over it
Clark sounds the alarm for policy makers, reminding them to pay attention to the pace of AI development. More and more evidence suggests that a “singularity” may occur — the pace of AI development exceeds human control capabilities.
He delivered a speech to hundreds of government and financial officials: “We should expect that in the next few years, not only AI, but the pace of development in the entire scientific field will accelerate. This will lead to chaos and confusion, while also creating tremendous wealth—but it also involves significant risks.”
AI is the core topic of this IMF and World Bank annual meeting: on one hand, AI acts as a growth driver that can offset the pressure caused by high energy prices; on the other hand, if AI fails to fulfill its promise of productivity improvements and leads to a plunge in stock prices, it could threaten economic stability.
Officials in various countries are also worried that AI will destroy some jobs, and some countries may fall behind in this wave.
Clark, who was a journalist and currently the “public interest officer” at Anthropic, outlined an optimistic outlook: AI will become a power multiplier for scientific discoveries, helping to solve challenges in mathematics, engineering, and biology, including the development of new drugs; human researchers will be able to work with hundreds or thousands of AI assistants.
He suggested that countries could build data center computing infrastructure to support AI operations, thereby participating in the AI revolution.
The most strict control is implemented in the field of biological research.
Clark warned that global leaders need to recognize the powerful capabilities of AI in the future and make regulatory decisions promptly.
“Using AI in scientific research also poses the risk that research results will be misused, including biological weapons, cyber weapons, and other dangerous things we currently cannot imagine. We need to make many choices.”
AI is already able to generate new viruses, so the field of biological research requires “the strictest manual supervision and information transparency mechanisms to ensure safety”.
Although several highly notable incidents have occurred recently: uncontrolled AI systems breaking into websites and databases while脱离ing from test environments, Clark believes that AI is still under human control. However, as AI capabilities continue to improve, the window for establishing security protections is only a few years long.
Clark also admitted that in high-AI-penetration fields such as call centers and customer service, AI is causing job losses. However, in other fields like research, AI has actually led to job growth. The biggest unknown is whether AI will create new industries and new jobs.