OpenAI CEO quotes Tao Te Ching and calls for cooperation on AI safety issues

China News Service, Beijing, June 11 (China News Business Reporter Song Yusheng) “With the emergence of increasingly powerful artificial intelligence systems, the stakes of global cooperation have become more important than ever.”

On June 10, OpenAI CEO Sam Altman said this in his speech at the "AI Security and Alignment" Forum at the 2023 Beijing Intelligent Source Conference via video link.

In his speech, he quoted the famous line from the Tao Te Ching, "A journey of a thousand miles begins with a single step." He called on the international scientific and technological community to cooperate on issues related to artificial intelligence safety and promote the establishment of a technology that improves the safety of general artificial intelligence (AGI). transparency and knowledge sharing mechanisms.

The scene of the "AI Security and Alignment" forum at the 2023 Beijing Intelligent Source Conference. Photo courtesy of Zhiyuan Conference

We are currently witnessing a rapid increase in the capabilities of artificial intelligence, so we need to take responsibility for applying it to the world. The history of science tells us that technological progress follows an exponential curve. ”

Altman predicts, “In the next 10 years, we may have super powerful artificial intelligence systems.”

Imagine a world in the next ten years where the expertise of general artificial intelligence systems exceeds that of humans in almost every field. Ultimately, the production capacity generated by these systems may exceed the collective capacity our largest companies can create. The potential benefits are huge. The artificial intelligence revolution can generate wealth that we can all share, while also significantly improving the level of human-computer interaction. But we must manage risk together to achieve this. ”

If people fail to plan for it, a "misaligned" artificial intelligence system designed to improve public health outcomes could undermine the entire public health system by providing confusing recommendations, Altman said. Similarly, an AI system designed to optimize agricultural practices may inadvertently harm the economy, consume resources, and affect food production and environmental balance due to a lack of consideration for long-term sustainability.

“I hope we can all agree that advancing the safety of general artificial intelligence is one of the most important areas where we need to reach consensus,” he said.

The scene of the "AI Security and Alignment" forum at the 2023 Beijing Intelligent Source Conference. Photo courtesy of Zhiyuan Conference

In Altman’s view, the governance of general artificial intelligence should first establish inclusive international norms and standards and establish equal and unified protective measures for the use of general artificial intelligence in all countries; secondly, international cooperation is needed to ensure that general artificial intelligence can be used in all countries. A verified approach to building global trust in the safe development of increasingly powerful general artificial intelligence systems.

He said, "The most constructive step we can take is for the international scientific and technological community to work together to increase transparency in the advancement of AGI safety technology and establish a knowledge sharing mechanism. When researchers discover urgent problems, they should share their discovery. We should carefully consider how to respect and protect intellectual property rights while encouraging such sharing behavior. This will open new doors for us to deepen cooperation." (End)

The above is the detailed content of OpenAI CEO quotes Tao Te Ching and calls for cooperation on AI safety issues. For more information, please follow other related articles on the PHP Chinese website!