Former Google CEO Eric Schmidt expressed shock at the rapid advancement of artificial intelligence in software development and declared himself obsolete after witnessing an AI system generate an entire program autonomously. It was a moment that crystallized both the promise and danger of the technology he had championed for decades.“That's terrible. I'm done,” Schmidt said at a forum at Harvard University earlier this month. “I've been programming for 55 years. It's really profound to see something start and end before your eyes in your life.”The confessions of one of Silicon Valley's most influential figures highlight a broader shift already underway, where AI is moving from a useful tool to a potential replacement for skilled human labor. Schmidt revealed that in major AI research labs, 10 to 20 percent of programming work is already done by AI systems, and that percentage is expected to increase rapidly in the coming years.
The bigger impact of AI is on automating business operations, not just code.
Schmidt argued that despite the dramatic demonstrations, AI is not only overrated but still overrated because the technology's true economic impact lies beyond coding. Speaking with Professor Graham Allison at Harvard University's John F. Kennedy Jr. Forum, he pointed to the automation of business operations as AI's greatest untapped potential.The real transformation is happening within companies, Schmidt explained, with AI systems taking over billing, accounting, product design, shipping and inventory management. These mundane processes quietly consume billions of dollars in corporate spending, and automating them has the potential to fundamentally reshape business operations.
Artificial general intelligence could be here by 2029, technology leaders warn
Schmidt predicted that artificial general intelligence (a system that rivals the smartest mathematicians, physicists, and artists) would emerge within three to five years. This timeline is driven by what he calls “recursive self-improvement,” where AI learns on its own without human guidance.“Computers are now improving themselves. They're learning how to plan so they don't have to listen to us anymore,” he warned at another recent event. Schmidt pointed out that AI is already writing 10 to 20 percent of the code at major research institutes such as OpenAI and Anthropic.Looking further ahead, Schmidt predicts that AI will enable autonomous learning capabilities within four years. “Many believe that next year we will have AI mathematicians,” he said, adding that the ability of computers to discover new facts and create medical breakthroughs seems imminent.However, Schmidt emphasized the need for human oversight as AI approaches these functions. “Someone is going to have to raise their hand and say, 'We've gone too far,'” he warned. “I believe there is no more important duty than to protect human agency and freedom.”A former Google executive suggested that Wall Street underestimated the magnitude of AI's impact on business automation and scientific discovery.
