It’s necessary ” pause “ the advance of artificial intelligence, claim in an open letter published on Tuesday March 28 a thousand industry experts and researchers, including Tesla CEO Elon Musk. They want to suspend for six months research on systems more powerful than GPT-4, the new language processing model launched in mid-March by OpenAI, the company behind the conversational robot ChatGPT − and that Elon Musk co-founded. There ” break “ should be used to develop better safeguards for such software, deemed “dangerous for humanity”.
“The past few months have seen AI labs locked in an uncontrolled race to develop and deploy ever more powerful digital brains, which no one – not even their creators – can reliably understand, predict or control”worry the signatories, with reference to the announcements of OpenAI and its partner Microsoft, but also those of Google or Meta, as well as many start-ups.
“Should we let the machines flood our information channels with propaganda and lies? Should we automate all jobs, including rewarding ones? Do we need to develop non-human minds that could one day be more numerous, smarter, and replace us? », ask the authors.
Among the signatories is one of the fathers of modern artificial intelligence, Yoshua Bengio, from the University of Montreal. But also figures from “tech” companies: the CEO of StabilityAI Emad Mostaque, as well as three researchers from Deepmind, a subsidiary of Google, or co-founders of Apple, Pinterest or Skype.
“Without any ethical control”
The text is also supported by several researchers who, for a long time, have sounded the alarm on the risks of AI, while hoping for its development: Max Tegmark, from the Future of Life Institute, which hosts the letter, Stuart Russel from l University of Berkeley, etc.
Several French people are signatories, including the specialist in artificial intelligence Raja Chatila, professor emeritus at the Sorbonne and member of the working group of the National Pilot Committee for Digital Ethics, to which the Minister Delegate for Digital, Jean-Noël Barrot, has asked for an opinion on the rise of programs like ChatGPT. “Everything that is happening in AI right now is happening without any real ethical or legal oversight. Companies are deploying tools on the web that have harmful effects,” explains who signed ” personally “. Also present are essayists critical of tech in general, such as Tristan Harris, from the Center for Humane Technology or Yuval Noah Harari, author of Sapiens.
You have 49.39% of this article left to read. The following is for subscribers only.