The White House has invited the leaders of companies that are highly advanced in artificial intelligence (AI) – Google, Microsoft, OpenAi and Anthropic – to come Thursday for a “frank discussion on the risks” associated with these technologies, with several members of the government, including Vice President Kamala Harris. “Our goal is to have a frank discussion on the current and short-term risks that we perceive in AI developments,” said the invitation consulted by AFP on Tuesday May 2. The government also wants to consider “steps to reduce these risks and other ways we can work together to ensure that the American people benefit from advances in AI while protecting them from harm.”
According to the White House, the four American bosses – Sam Altman for OpenAI, Dario Amodei for Anthropic, Satya Nadella for Microsoft and Sundar Pichai for Google – have confirmed their participation. President Joe Biden “clearly” said last month that these companies “must make sure their products are safe before making them available to the general public,” the invitation says. A White House official said the government wanted to emphasize the need to “innovate in a responsible, ethical and trustworthy manner.” Interviews are also underway with various researchers, companies and NGOs to feed the reflection on AI.
This technology has been very present in everyday life for years, from social media recommendation algorithms to recruitment software and many high-end household appliances. But the runaway success this winter of ChatGPT, the generative AI interface of OpenAI, a largely Microsoft-funded start-up, has sparked a race for ever more intuitive and capable systems that is generating excitement and concern from a new magnitude. Especially when Sam Altman, the boss of OpenAI, talks about the upcoming advent of so-called “general” AI, when programs will be “smarter than humans in general”. Geoffrey Hinton, considered one of the founding fathers of AI, warned Monday of “profound risks to society and humanity” in an interview with The New York Times after resigning from his post at Google.
On Monday, IBM boss Arvind Krishna told Bloomberg that he was considering drastically reducing the IT giant’s administrative staff, given the potential for artificial intelligence (AI) and automation technologies to run this job. type of tasks. “It seems to me that 30% (of the 26,000 administrative employees) could easily be replaced by AI and automation over a five-year period,” he said.
In terms of regulation, Europe hopes to again lead the way with an ad-hoc regulation, as it did with the law on personal data. The White House released a “Plan for an AI Bill of Rights” in late 2022, a brief document that lists general principles such as protection against dangerous or fallible systems.