Baidu CEO Robin Li shared his outlook on the future of artificial intelligence (AI), asserting that the industry is experiencing a “bubble” that will eventually lead to a significant shakeout. Speaking at the Harvard Business Review Future of Business Conference, Li estimated that only about 1% of AI companies will survive, as others are likely to disappear when the AI “bubble” inevitably bursts.
AI Hallucinations “Solved,” Says Baidu CEO
Li expressed confidence in the improvements to large language models (LLMs), noting that they now produce more accurate responses than before. “The most significant change we’re seeing over the past 18 to 20 months is the accuracy of those answers from large language models,” he said, stating that hallucinations—a tendency for AI to produce false or misleading information—are largely resolved. Li believes that users can now “basically trust the answer” when engaging with advanced chatbot models.
AI Sector and the Dot-Com Bubble Parallel
Comparing the current state of AI to the dot-com bubble of the 1990s, Li projected that the AI field would undergo a similar phase of overvaluation followed by a steep correction, ultimately resulting in a small number of dominant companies. “Probably one percent of the companies will stand out and create tremendous value for society,” he noted, suggesting that a consolidation phase is inevitable as the industry matures.
Preparing for Long-Term Technological Shifts
Despite the forecasted downsizing, Li does not expect AI to eliminate jobs in the near future. He estimated that it could take another 10 to 30 years for AI advancements to start replacing human jobs on a large scale. He emphasized the importance of preparing for this shift, encouraging organizations, governments, and individuals to ready themselves for a world where AI will have a more profound role.
Li’s comments come amid rapid advancements and growing investment in AI across industries, with companies racing to incorporate AI into services, products, and operations. As major tech firms, including Baidu, continue to pour resources into AI, Li’s cautionary outlook reflects both optimism for AI’s potential and a warning of the challenges ahead for new market entrants.
