Sam Altman, CEO of OpenAI, has admitted he was wrong about the rapid economic disruption his company's AI technology would cause. He had predicted widespread upheaval and the immediate disruption of software businesses following the release of GPT-4 in 2023. However, Altman conceded on David Senra's Founders podcast that society and the economy are adapting much more slowly than anticipated. He attributed this inertia to people continuing their established routines and purchasing from familiar vendors. Altman even admitted to personally preferring his old workflows over new AI coding tools. He views this slower adaptation as a positive, suggesting it will lead to a smoother transition and mitigate job losses. This admission validates concerns from market analysts who warned that massive capital expenditures in AI might be outpacing real-world adoption. The AI bubble investment is now being compared to historical manias like the dot-com bubble. In response to potential softening enterprise demand, Nvidia is reportedly positioning itself as a buyer of last resort for its own GPUs, investing in AI startups to create internal demand. Furthermore, Altman criticized AI safety advocates, particularly Anthropic CEO Dario Amodei, for what he called "doomsday marketing" and a "benevolent dictator" narrative. He argued that concentrating unchecked authority in the hands of tech elites, under the guise of existential risk, is a greater danger than rogue AI. This shift in focus allows Altman to pivot away from discussing capital cycles and his overstated timelines.