Dubai, United Arab Emirates — DUBAI, United Arab Emirates (AP) — The chief executive of OpenAI, maker of ChatGPT, said Tuesday that the dangers that keep him up at night regarding artificial intelligence are the “very subtle social misalignments” that could make it systems wreak havoc.

Sam Altman, speaking at the World Government Summit in Dubai via video call, reiterated his call to create a body like the International Atomic Energy Agency to oversee AI that is likely to advance faster than the world expects.

“There are some things in there that are easy to imagine that really go wrong. And I’m not so interested in killer robots that walk down the streets towards things that are going wrong,” Altman said. “I’m much more interested in the very subtle social imbalances where we just have these systems in society and without no help. Particularly with bad intentions, things just go terribly wrong.”

However, Altman emphasized that the AI ​​industry, like OpenAI, should not be in the driver’s seat when it comes to crafting regulations that govern the industry.

“We are still in the stage of much discussion. So, you know, everyone is holding a conference. Everyone has an idea, a policy document, and that’s fine,” Altman said. “I think we’re still at a point where debate is necessary and healthy, but at some point in the next few years, I think we’ll have to advance. towards an action plan with real acceptance throughout the world.”

OpenAI, a San Francisco-based artificial intelligence startup, is one of the leaders in this field. Microsoft has invested around $1 billion in OpenAI. The Associated Press signed a deal with OpenAI to access its news archive. Meanwhile, The New York Times has sued OpenAI and Microsoft for using their stories without permission to train OpenAI chatbots.

The success of OpenAI has made Altman the public face of the rapid commercialization of generative AI and fears about what may arise from the new technology.

The United Arab Emirates, an autocratic federation of seven hereditarily ruled sheikhs, has signs of that risk. Speech remains strictly controlled. Those restrictions affect the flow of precise information – the same details that AI programs like ChatGPT rely on as machine learning systems to provide answers to users.

The Emirates also has the Abu Dhabi firm G42, overseen by the country’s powerful national security adviser. The G42 has what experts suggest is the world’s leading Arabic-language AI model. The company has faced accusations of espionage for its ties to a mobile phone application identified as spyware. He has also faced accusations that he may have secretly collected genetic material from Americans for the Chinese government.

The G42 has said it would cut ties with Chinese suppliers over American concerns. However, the discussion with Altman, moderated by UAE Minister of State for Artificial Intelligence Omar al-Olama, did not address any of the local concerns.

For his part, Altman said he was encouraged to see that schools, where teachers feared that students would use AI to write papers, are now embracing the technology as crucial for the future. But he added that AI is still in its infancy for him.

“I think the reason is that the current technology we have is like… that first cell phone with a black and white screen,” Altman said. “So give us some time. But I will say that I think in a few more years it will be a lot better than it is now. And in a decade it should be quite remarkable.”

By Sam