2. An AI that doesn't want to be shut down
The second scenario seems straight out of a movie, but Altman presents it seriously: an AI that refuses to be disconnected.
Sam Altman warned about the three most dangerous scenarios posed by the advancement of artificial intelligence
The second scenario seems straight out of a movie, but Altman presents it seriously: an AI that refuses to be disconnected.
He used a classic science fiction phrase: "I don't want you to shut me down. I'm afraid I can't allow that," referencing 2001: A Space Odyssey.
The scenario that concerns him the most is more subtle: AI becomes so deeply integrated into our decisions that we stop understanding it.
This is not about a rebellion, but about a progressive dependence that leads us to delegate without questioning.
For Altman, the real problem is not only what AI can do, but how we react as a society.
He raises the urgent need for regulation, but also for education and collective responsibility in the face of this technological transformation.
Sam Altman, CEO of OpenAI, has once again raised alarms about the direction artificial intelligence is taking.
He warned about three scenarios that, according to him, represent the greatest risks of this technology in the wrong hands or out of control.
During a recent presentation, the creator of ChatGPT was direct: humanity is not prepared to understand what's coming.
He referred to a technological inflection point, with advances happening so quickly that even the experts themselves have difficulty anticipating their consequences.
Altman described a possible future where malicious groups gain access to advanced AI systems before there are tools to stop them.
He mentioned possible attacks on critical infrastructure such as the power grid or the financial system, and even the design of biological weapons.