CEOs of leading technology companies called for slowing down the development of artificial intelligence: reasons

Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman have advocated slowing down the development of the most powerful artificial intelligence systems. They believe this is necessary to strengthen security and create mechanisms to control the development of the technology.

This is discussed in the articles We Must Pace the Frontier and Bloomberg .

Amodei called for a slowdown in the development of the most powerful AI systems so that security experts can keep up with their growing capabilities. He emphasized that this is not about completely stopping progress, but about a more controlled development of the technology.

One of the concerns, the Anthropic CEO said, is the recursive self-improvement of AI, where artificial intelligence is used to create and improve subsequent generations of models. According to him, this process can outpace the ability of developers to understand and control such systems.

Amodei also proposed involving independent external experts in the work of leading AI companies, establishing common security standards, and strengthening international coordination.

He believes that an additional 1-2 years before the emergence of AI with critically high capabilities could give researchers time to improve security methods and test the behavior of models.

In parallel, OpenAI CEO Sam Altman told company employees that OpenAI could slow down the development of advanced AI systems by potentially coordinating such actions with other developers.

OpenAI Chief Scientist Jakub Paczocki also said that AI companies should coordinate to slow down future development as needed until common security standards are established.

According to Bloomberg, OpenAI has already slowed down some stages of model development and recently suspended some internal AI training due to security concerns.

At the same time, Altman and Amodei do not advocate a complete rejection of the development of artificial intelligence. They are talking about controlling the pace of development of the most powerful systems and strengthening security measures.

Recall that in the US, an artificial intelligence-based facial recognition system wrongly accused Tennessee resident Angela Lipps of bank fraud. As a result, the woman spent almost 6 months in prison until evidence proved her innocence.

Also follow “Pryamim” on Facebook , X , Telegram , and Instagram.