Tech Leaders Predict Superintelligent AI, Experts Offer Caution and Context

Bold predictions from Big Tech about the arrival of superintelligent artificial intelligence have reanimated a familiar debate about timing, risk and readiness. At the World Economic Forum in Davos this year Anthropic CEO Dario Amodei stated, "By 2026 or 2027, we will have AI systems that are broadly better than almost all humans at almost all things," a forecast that sits at the more optimistic end of public statements from industry leaders. Elon Musk has argued we will see systems more intelligent than any single human within months and will "100%" have an AI exceeding the intelligence of all humans combined by 2030. OpenAI CEO Sam Altman told Bloomberg he thinks AGI will "probably get developed" before the end of the next presidential term.

Those high profile assertions are contrasted by the findings of a study reported by Gizmodo that surveyed a broader set of AI experts. The study found most participants disagreed with the view that AI will accelerate at "light speed" to reach superintelligence by the end of the decade. Still, the aggregate assessment among experts was not complacent. Respondents predicted that AI will become one of the defining technologies of the coming decades, likening its societal role by 2040 to that of electricity in transforming industry, daily life and institutions.

The study offered concrete projections for the nearer term. By 2030 experts expect AI to provide daily companionship for roughly 15 percent of adults and to assist in approximately 18 percent of work hours in the United States. Those figures underscore a likely reconfiguration of labor and social life even if machines do not attain the dramatic threshold of general intelligence in the next few years.

The divergence between industry leaders and the expert median reflects both forecasting uncertainty and differing incentives. Company executives face competitive pressures and investor expectations that reward bold visions of rapid progress. Academic and independent researchers often emphasize methodological challenges, from reproducibility to generalization across tasks, and the long history of over optimistic technological timelines. Forecasting AI trajectories is especially fraught because development depends on both technical breakthroughs and the allocation of talent, compute and capital.

Policy makers and institutions are now confronted with a dual task. They must prepare for significant, tangible changes to employment, mental health and public services stemming from advanced AI applications, while also crafting governance mechanisms for the more extreme scenarios that industry leaders highlight. That means accelerating workforce retraining, updating safety and liability frameworks, and investing in research that evaluates systemic risk without stifling beneficial innovation.

Ethical choices will matter. How societies distribute the gains from highly capable AI, protect privacy and autonomy, and ensure transparency in decision making will determine whether the technology amplifies prosperity or inequality. The current debate is less about whether AI will be transformational and more about when and under what safeguards that transformation will arrive. As the conversation shifts from prediction to preparation, the challenge for leaders and regulators will be to balance ambition with rigorous assessment and public accountability.