Predict Altman’s poison for the future of artificial intelligence agents and its challenges

Sam Altman, CEO of Openai, predicts a future in the TED2025 event in which independent artificial intelligence agents operate on the Internet, accessing sensitive users’ information such as bank accounts and may even take control of systems; An attractive but full of security challenges that makes users doubt.

According to the WindowsCentral report, artificial intelligence tools such as ChatGPT have been expanding in recent years. However, concerns about security and privacy remains a major obstacle to the full acceptance of the technology.

Factor -based artificial intelligence errors can be very costly

In an interview at TED2025, Sam Altman, CEO of Openai, addressed the future challenges of the technology, pointing out that the AI ​​-based artificial intelligence that can operate automated and independently has a likely but risky future. She said.

“I think people in many cases will be accustomed to the agent -based artificial intelligence, but even if some are comfortable with the technology and some, we will finally go into the stage where artificial intelligence systems will click on the Internet. “This is one of the most interesting and most popular safety challenges we have encountered, because when artificial intelligence has access to systems and information and makes a mistake, its mistakes will have much more risks.”

Altman emphasized that if users do not trust artificial intelligence systems, they will never allow them to access personal information or control their systems:

“Good product is a safe product. “You will not use them until our artificial intelligence (artificial intelligence) do not empty your bank account or will not delete your data.”

Microsoft faced major security problems in introducing the Windows Recall tool. Even when the sensitive information filter was enabled, it still recorded important data such as bank card information. This challenged public confidence in artificial intelligence systems with extensive access.

Overall, the future of operating artificial intelligence, despite the significant potential, still faces serious challenges in security and privacy. Sam Altman’s remarks on widespread access to personal information such as bank accounts, along with examples such as Microsoft’s Windows Recall Windows Recall tool, show that users’ trust requires strong security guarantees and performance transparency. Although the path to this emerging technology seems to be clear and rapid, its success will depend on the ability of companies to manage risks and improve systems safety.