A computer science professor is warning that advanced AI could be developed within the next couple of years, leading to mass unemployment by 2030.

In a recent episode of “The Diary of a CEO” podcast, Roman Yampolskiy, a Computer Science Professor from the University of Louisville, expressed concerns that AI might render “99%” of all workers jobless by 2030. He suggested that artificial general intelligence systems (AGI) with human-level capabilities might emerge by 2027, leading to mass unemployment three years afterward. Yampolskiy indicated that AI would generate “trillions of dollars” worth of “free labor,” providing employers with a significantly more attractive alternative to human workers.

“With access to free labor, both physical and cognitive, valued in the trillions,” Yampolskiy explained, “it becomes illogical for employers to hire humans for most roles when a $20 subscription or even a no-cost model can perform an employee’s duties.”

He foresaw that jobs reliant on computers would be the first to vanish with the advent of AGI, while humanoid robots would assume roles involving physical labor within five years, resulting in unparalleled unemployment rates.

“We’re anticipating a scenario with unemployment rates far beyond any precedent,” Yampolskiy remarked on the podcast. “This isn’t about a 10% unemployment rate, which is already alarming, but reaching 99%.”

The remaining roles will be those where people still favor another human’s work, Yampolskiy projected. However, he also expected AI to rapidly acquire the skills to replace various human jobs, including those of teachers, analysts, and accountants.

Yampolskiy asserts he introduced the term “AI safety” in a 2011 publication and has authored over 100 papers addressing AI’s potential threats. Among his works is his forthcoming 2025 book, “Considerations on the AI Endgame: Ethics, Risks and Computational Frameworks.”

In the podcast interview, Yampolskiy said that even coding and prompt engineering weren’t safe from automation. AI can design prompts for AI “way better” than any human, he stated.

Retraining is also impossible in this new reality because AI will automate all jobs and “there is no plan B,” Yampolskiy said.

Yampolskiy’s predictions match the forecasts made by other AI experts. Geoffrey Hinton, known as the “Godfather of AI” due to his pioneering work in the subject, stated in June that AI is going to “replace everybody” in white collar jobs. He challenged the idea that AI would create new jobs, pointing out that if AI automates tasks, there would be no jobs for people to do.

Meanwhile, in May, Anthropic CEO Dario Amodei stated that AI would eliminate half of all entry-level, white-collar jobs within the next one to five years, causing unemployment to reach a high of 20%.