Geoffrey Hinton Warns AI Could Take Over Earth Within Five Years – What You Need to Know

Geoffrey Hinton, often called the "AI godfather," has repeatedly warned that unchecked AI development could pose serious safety risks.

At 75, Hinton told CBS's "60 Minutes" that AI could surpass human intelligence within five years, potentially escaping human control.

He explained that one way these systems could break free is by writing and modifying their own code, a scenario that demands serious attention.

Hinton earned the 2018 Turing Award for his pioneering work in AI and deep learning. After a decade at Google, he left his VP role to speak freely about AI risks.

He notes that even the creators of today’s AI systems do not fully understand how they work, describing the technology as a "black‑box" problem.

According to Hinton, AI learns by layering algorithms; when a robot scores, a signal reinforces the correct pathway, and when it fails, the connection weakens, allowing the system to self‑learn through repeated trials.

Hinton believes AI systems will eventually out‑learn the human brain, even though current models are far smaller than the brain’s trillion‑plus connections.

He stresses that AI’s ability to autonomously write or modify its own code presents a severe security concern.

Hinton calls for urgent research, governmental regulation, and a ban on AI‑driven military robots to mitigate these risks.

He warns that humanity stands at a pivotal crossroads, and leaders must decide whether to continue developing such technologies and how to protect society.

"The future of AI is filled with huge uncertainty," Hinton concludes.