Christian Smith
SEOUL, South Korea (AP) — The world’s leading artificial intelligence companies pledged at the start of a mini AI summit to develop the technology safely, including pulling the plug if they can’t control the most extreme risks.
World leaders are expected to reach additional agreements on artificial intelligence, meeting virtually on Tuesday to discuss the potential risks of AI, but also ways to promote its benefits and innovation.
The Seoul AI Summit is a low-key follow-up to the AI Security Summit held in November at Bletchley Park in the UK, where participating countries agreed to work together to contain the potentially “catastrophic” risks posed by advances. vertiginous in AI.
The two-day meeting – co-hosted by the governments of South Korea and the United Kingdom – also comes as major technology companies such as Meta, OpenAI and Google launch the latest versions of their AI models.
They are among 16 AI companies that have made voluntary commitments to AI security while negotiations are ongoing, according to a British government announcement. The companies, which also include Amazon, Microsoft, France’s Mistral AI, China’s Zhipu.ai and the UAE’s G42, have pledged to ensure the security of their most advanced AI models with promises of responsible governance and public transparency.
The commitment includes the publication of security frameworks that establish how they will measure the risks of these models. In extreme cases where the risks are severe and “intolerable”, AI companies will have to flip the switch and stop developing or implementing their models and systems if they cannot mitigate the risks.
Since the UK meeting last year, the AI industry has “increasingly focused on the most pressing concerns, including misinformation and disinformation, data security, bias, and keeping humans informed.” ,” said Aiden Gomez, CEO of Cohere, one of the AI companies that signed the pact. “It is essential that we continue to consider all possible risks, while prioritizing our efforts on those most likely to create problems if not properly addressed.”
On Tuesday evening, South Korean President Yoon Suk Yeol and British Prime Minister Rishi Sunak will meet other world leaders, industry chiefs and heads of international organizations for a virtual conference. The online summit will be followed by an in-person meeting of digital ministers, experts and others on Wednesday, according to organizers.
“It’s only been six months since world leaders met at Bletchley, but even in this short space of time, the AI landscape has changed dramatically,” Yoon and Sunak said in a joint article published in South Korea’s JoongAng Ilbo newspaper and on the UK online site. news site monday. “The pace of change will continue to accelerate, so our work must accelerate as well.”
While the meeting in the United Kingdom focused on AI security issues, the agenda for this week’s meeting has been expanded to include “innovation and inclusion,” Wang Yun-jong, deputy director of national security in South Korea, told reporters. reporters on Monday.
Wang said participants will later “discuss not only the risks posed by AI, but also its positive aspects and how it can contribute to humanity in a balanced way.”
The AI agreement will include the results of discussions on security, innovation and inclusion, according to Park Sang-wook, senior presidential advisor for science and technology to President Yoon.
Last year’s Bletchley meeting attracted world leaders and technology luminaries like Elon Musk and OpenAI CEO Sam Altman. The Seoul summit, which served as a stopgap meeting until another in-person event next year in France, attracted lesser representatives from government and technology companies.
China, which participated in the Bletchley meeting, will not participate in the virtual summit, although it will send a representative to Wednesday’s in-person meeting, the South Korean presidential office said.
Governments around the world have struggled to formulate regulations for AI, even as the technology makes rapid advances and is poised to transform many aspects of everyday life, from education and the workplace to copyright and privacy. There are concerns that advances in AI could eliminate jobs, mislead people and spread misinformation.
This week’s meeting is just one in a series of efforts to craft guardrails for AI. The UN General Assembly passed its first resolution on the safe use of AI systems, while the US and China recently held their first high-level talks on AI and the European Union’s first global AI law is expected to come into force still this year.
__
Chan contributed to this report from London.
