Openai CEO Sam Altman has expressed growing concern over the misuse of artificial intelligence, warning that the actual dangers do not lead to autonomous machines being fraudulent, but rather that they use AI tools to cause intentional harm.
Speaking about a recent episode of Theo von's podcast, Altman tackled a long-standing question of AI risk. He shifted the spotlight to human intentions rather than reflecting the dystopian fears of mechanical opposition to humanity. “People who use AI to do bad things are more worried than AI decides to do bad things themselves,” says Altman.
His remarks show a departure from the typical science fiction narrative of killer robots and self-aware systems, instead highlighting more immediate and realistic challenges. This highlights the possibility that malicious actors will utilize advanced AI models.
“The risk is when someone really wants to do harm and has a very powerful tool to do that,” he pointed out, noting that if a powerful AI system is not checked, the weapon can be made easier.
Altman acknowledged the difficulty of designing AI systems that remain safe and beneficial while in the hands of millions of users. “We're trying to build guardrails as we go. It's difficult, but it's necessary,” he admitted, highlighting the ongoing efforts at Openai to incorporate ethical guidelines and technical protection measures into its model.
His comments come when Openai faces an increase in scrutiny from policymakers and civil society, especially as speculation occurs around the development of the GPT-5. As generational AI is accessible and influential in everyday life, questions about governance, accountability and control are more pressing than ever before.
Meanwhile, Openai officially launched its deployment of its new artificial intelligence agent, ChatGpt Agent, after a week-long delay. According to a statement posted on social media platform X, the feature, originally announced on July 18th, is now available to all ChatGpt Plus, Pro and team subscribers.
The rollout delay has confusing many users, and some have reported that despite Openai's claims of full deployment, it is not featured. The company has not revealed the cause behind the delay, and questions raised in the comments section of the post remain unanswered.
