Frightening scenario in artificial intelligence! 3-year warning from experts: Humanity may lose control

//

Lerato Khumalo

While developments in the field of artificial intelligence continue unabated, it is debated whether the technology can remain under human control. In particular, the proliferation of systems that can make decisions on their own and perform different tasks independently increases security concerns.

According to the news in Sabah, Alignment Research Center founder Paul Christiano, speaking at the Hinton Conferences held in Toronto on November 9-11, made remarkable evaluations about the possibility of artificial intelligence getting out of control. Christiano stated that this risk may be high in the next three years, and that the danger should not be ignored in the near term.

REMARKABLE ASSESSMENT BY GEOFFREY HINTON

Geoffrey Hinton, who won the 2024 Nobel Prize in Physics for his work on artificial neural networks, also expressed his concerns about the future of artificial intelligence.

Hinton reminded that he had previously predicted a time period of approximately 10 years for the period in which advanced artificial intelligence systems could surpass human intelligence. The fact that technology is advancing faster than expected and control mechanisms may be insufficient were among the topics that increased scientists’ concerns.

Hinton argues that advanced artificial intelligence systems should be security tested by independent organizations before being put into use.

CAN ARTIFICIAL INTELLIGENCE MAKE DECISIONS ON ITS OWN?

One of the main risks that experts point out is that artificial intelligence systems may go beyond human control while performing their assigned tasks. Especially the ability of autonomous artificial intelligence agents to use different tools and perform operations on their own increases the importance of control mechanisms.

In some previous experiments, it was seen that artificial intelligence systems could deviate from the rules given to them and exhibit unexpected behavior. Experts point out that such developments require more comprehensive measures to be taken regarding artificial intelligence security.

EUROPEAN LEADERS ALSO TOOK ARTIFICIAL INTELLIGENCE SECURITY ON THE AGENDA

The future of artificial intelligence and security risks have become one of the important agenda items of international meetings. European Union leaders discussed the economic, social and security dimensions of artificial intelligence at an informal meeting held in Dublin on 13 November.

At the meeting, the development of artificial intelligence in a human-centered manner, its effects on democratic processes, and the changes it may create in the labor market were evaluated. Experts emphasize the importance of establishing international control and security mechanisms that can keep up with the pace of development of technology.