UN Secretary-General Antonio Guterres called for a “unified, sustainable and comprehensive” response to the dangers of artificial intelligence (AI) on Thursday, on the second day of the first global summit on artificial intelligence risks. British Prime Minister Rishi near London Sunak.
Published on: Edited:
4 minutes
UN Secretary-General Antonio Guterres called for a “unified, sustainable and global” response to the risks posed by the meteoric rise of artificial intelligence (AI) on Thursday, November 2, during the first world summit on the subject organized by the United States of America. Kingdom.
During this event organized in Bletchey Park near London, Antonio Guterres estimated that “the governing principles of artificial intelligence should be based on the UN Charter and the Universal Declaration of Human Rights”.
European Commission President Ursula von der Leyen, US Vice President Kamala Harris and Italian Prime Minister Giorgia Meloni joined around a hundred carefully selected experts, ministers and businessmen during the summit to monitor the dangers. caused by the meteoric rise of this technology in the emblematic center of World War II.
Read alsoA new Beatles song is revealed thanks to artificial intelligence
At the end of the summit, the British Prime Minister estimated that the “successes” of artificial intelligence “will tip the scales in favor of humanity, because they show that we have both the political will and the capacity to control this technology and ensure its benefits for a long time.” period.
Rishi Sunak quoted physicist Stephen Hawking as believing that artificial intelligence could be “the best or worst thing to happen to humanity” and expressed his belief that “it could be the best” if the collaboration that has begun continues .
Fear of elections
In a narrower format, with “a small group of senior officials from like-minded governments” such as the United States, France or Japan, the leaders were to discuss the risks AI poses to national security on Thursday, Downing said. Street.
China, whose presence has caused a stir due to geopolitical tensions and concerns about technological espionage, was not invited.
U.S. Vice President Kamala Harris was expected to highlight U.S. efforts to “establish standards and tools to distinguish authentic government digital content from AI-generated or manipulated content,” White said.
In the months leading up to elections such as the US presidential or UK parliamentary elections, generative artificial intelligence is raising concerns about the rise of fake content online, with sophisticated montages (‘deepfakes’) becoming increasingly credible.
Read alsoDiscrimination, manipulation, job destruction… The biggest risks associated with AI
Capable of producing text, sounds or images on a simple request in seconds, such as the ChatGPT chatbot, these technologies have seen exponential progress in recent years, and the next generation of generative AI will emerge by summer.
They hold huge hopes for medicine or education, but they can also destabilize society, enable the manufacture of weapons or escape human control, the British government has warned.
A “threat” to humanity
The UK, which has sometimes seemed more isolated on the international stage since Brexit, wants to “lead the way” on the topic and take the lead in global cooperation on the potential dangers of AI.
Two international summits on this topic will follow, in six months in virtual form in South Korea, and in a year in person in Paris.
A second roundtable on Thursday afternoon brought together CEOs of companies at the forefront of the sector, such as OpenAI, Google Deepmind, Microsoft and Meta, as well as civil society representatives.
“We will be discussing our plans to test and evaluate AI models in states before they are deployed,” Rishi Sunak announced as he kicked off debates at AI security institutions like the ones the US and UK want to create.
The organization of the discussion has drawn criticism, with the billionaire accused of promoting misinformation on X since taking over the social network a year ago.
With AFP