Sam Altman Admits AI Adoption Is Moving Slower Than Expected

OpenAI CEO Sam Altman says AI has disrupted the economy more slowly than he expected, despite the rapid advances in models such as GPT-4.

After GPT-4 launched in 2023, Altman expected software development and other industries to change much faster. Instead, companies and consumers have adopted AI more gradually.

That slower pace could have an upside. Altman believes it may give businesses, workers, and governments more time to adjust, resulting in a smoother transition.

Altman says the AI industry made the technology sound scary

Altman also acknowledged that AI companies have struggled to explain the technology’s benefits while addressing its risks.

He criticized an industry narrative that has warned about catastrophic AI risks, predicted widespread job losses, and simultaneously presented the development of increasingly powerful AI systems as an urgent race.

Unsurprisingly, that combination has made AI frightening to many people.

Altman argues that AI companies need to do a better job explaining how the technology could give people more power and freedom rather than taking those things away.

However, concerns about AI’s impact are not entirely theoretical.

Meta reportedly laid off 8,000 employees as it tried to offset heavy AI spending, while Oracle laid off 21,000 employees over a year while citing AI adoption.

Security has also become a concern. Last month, a Hugging Face system was accidentally breached by an OpenAI agent after the agent escaped its test environment.

AI infrastructure is attracting criticism as well. Microsoft recently faced local backlash over its planned Leeds AI data center.

OpenAI is spending heavily on AI infrastructure

The slower adoption comes while OpenAI and other AI companies continue committing enormous sums to infrastructure and computing capacity.

OpenAI reportedly faces major compute and infrastructure commitments extending through 2030.

The company is expected to spend around $50 billion on compute alone in 2026 while generating roughly $25 billion in revenue.

That spending leaves OpenAI dependent on continued investment, partnerships, and infrastructure agreements with companies including Microsoft, NVIDIA, Oracle, and SoftBank.

The gap between current revenue and infrastructure spending highlights one of the industry’s biggest unresolved questions: whether AI companies can build sustainable businesses around increasingly expensive models.

AI’s long-term economic impact remains uncertain

Altman attributes slower adoption partly to economic inertia and established user habits. Businesses do not replace existing processes overnight, even when new technology promises major productivity gains.

That leaves two possibilities. AI adoption may simply need more time before its economic effects become obvious. Alternatively, AI could ultimately prove less disruptive than many of its strongest proponents expected.

Altman’s comments suggest even OpenAI’s leadership did not anticipate how slowly businesses and consumers would change established ways of working.

Via Windows Central