The British government announced that representatives of 16 leading companies in the field of artificial intelligence in the world - in the South Korean capital on Tuesday - have pledged new commitments to ensure the safe development of this science.
British Prime Minister Rishi Sunak said in a statement from the British Institute of Science, Innovation and Technology: "These commitments ensure that the world's leading AI companies are committed to transparency and accountability about their plans to develop secure AI."
The agreement, signed in particular by OpenAI (ChatGPT), Google DeepMind and Anthropic, builds on the consensus reached at the first global "summit" on AI security last year in Bletchley Park in the United Kingdom.
Seoul is hosting the second edition of the summit in partnership between the South Korean and British governments.
"Unforgiven" risks
AI companies, which have yet to announce how they assess the security of the systems they develop, are obliged to identify the risks that "cannot be overlooked" and the steps they will follow to ensure these ceilings are not exceeded, according to the statement.
These ceilings will be set ahead of the next AI "summit" in 2025 in France.
Among the companies adhering to these security rules are US tech giants Microsoft, Amazon, IBM, Meta, French company Mistral (artificial intelligence), and Zhopio. AI (Zhipu.ai) Chinese.
The two-day Seoul summit will be held, in part, via videoconferences.
In addition to addressing the security aspect of AI systems, participants will examine ways in which governments can foster innovation and expand AI research in universities.
Informative
ReplyDeleteGreat benchmarks set for responsible AI development.
ReplyDeleteInformative
ReplyDelete