Archive · 10 Sep 2026
AI Could Threaten Humanity, Godfather of AI Warns
Geoffrey Hinton is a famous AI scientist who worries about very powerful artificial intelligence.
He said there may be more than a 10% chance that AI could cause humans to disappear within the next ten years.
He also said this number is only an estimate because nobody knows exactly what will happen.
People have never created machines that might become smarter than them.
Hinton said AI would not need a robot body to cause serious harm.
It could influence people by talking to them or help create dangerous viruses.
It might also be used for major computer attacks.
Hinton believes scientists should work on making AI avoid harmful goals.
He says the risk could be much higher than 1%, but no one can calculate it precisely.
Geoffrey Hinton said a 10% chance of AI causing human extinction within a decade is not unreasonable.
Hinton stressed that nobody knows how to estimate the risk reliably because humanity has never created potentially smarter beings.
He warned that AI could cause devastation through manipulation and conversation without controlling the physical world.
Potential dangers include designing biological or computer viruses and launching large-scale cyberattacks.
Hinton said advanced AI should be designed so it does not develop harmful goals.
- Who
- Geoffrey Hinton, an artificial intelligence pioneer and Nobel Prize winner, discussed the risk.
- What
- Hinton said a 10% chance of AI causing human extinction is not an unreasonable estimate.
- When
- During an interview with the BBC; the article does not specify the date.
- Where
- In a BBC interview with journalist Victoria Derbyshire.
- Why
- Hinton is concerned that advanced AI could manipulate people, design dangerous viruses, launch cyberattacks, or create other unforeseen risks.
- Risk estimate
- Hinton said a 10% chance of AI causing human extinction is not unreasonable.
- Time frame
- The discussion concerned the possibility of extinction within the next decade.
- Uncertainty
- Hinton said nobody knows how to make a sensible or reliable estimate.
- Potential mechanism
- AI could cause chaos by talking to and influencing people.
- Potential threats
- Hinton cited biological and computer viruses, large-scale cyberattacks, and other unforeseen dangers.
- Proposed safeguard
- He said advanced AI should be designed not to develop harmful goals.