Creator of AI: We Have 2 Years Before Everything Changes! These Jobs Won't Exist in 24 Months! - The Diary of a CEO with Steven Bartlett Recap

Podcast: The Diary of a CEO with Steven Bartlett

Published: 2025-12-18

Duration: 1 hr 40 min

Guests: Yoshua Bengio

Summary

Yoshua Bengio, an AI pioneer, warns of the potential dangers of AI, including job loss and security risks, emphasizing the need for robust regulation and ethical AI development.

What Happened

Yoshua Bengio, a prominent AI pioneer, expresses urgent concerns about the rapid advancement of AI and its potential to develop goals beyond human control. He warns that AI's increasing capabilities, such as autonomous weapons, pose a threat to human safety and could lead to catastrophic outcomes if not properly regulated.

Bengio highlights the democratization of dangerous knowledge through AI, such as the creation of weapons, which could fall into the wrong hands. He notes that while the probability of such risks might be low, the severity of possible outcomes demands precautionary measures similar to those in other high-stakes fields.

He argues that AI's rapid evolution could lead to significant job losses, particularly in cognitive roles, within the next five years. The advancement of robotics, driven by cheaper AI software, further accelerates this trend, with companies like Elon Musk's Tesla pivoting towards humanoid robots.

Bengio stresses the importance of public opinion in shaping AI policy, drawing parallels to how public awareness influenced nuclear policy during the Cold War. He believes that raising awareness about AI risks can lead to more responsible development and regulation.

The discussion also touches on AI's potential to concentrate power in the hands of a few corporations or countries, which could exacerbate inequality and threaten democratic systems. Bengio advocates for distributed power and technical solutions to ensure AI systems are aligned with human values.

Despite the risks, Bengio remains hopeful, citing a belief in the possibility of developing AI that can benefit humanity without causing harm. His organization, LawZero, focuses on creating safer AI training methods as a step towards this goal.

Key Insights