ROME (Reuters) – An Italian senator said he read a speech made by a chatbot to parliament before performing a stunt to provoke a “serious debate” about the risks and opportunities posed by artificial intelligence, calling his colleagues Surprised me.
“How many people today can distinguish between a text generated by human intelligence and a stream of thought generated by an artificial intelligence algorithm?” centrist Senator Marco Lombardo asked his colleagues. .
The senator told Reuters on Thursday that his speech on the subject of various bilateral agreements with Switzerland was produced by an open AI GPT-4 chatbot backed by Microsoft.
He said it took “several hours” for staff to enter the required information into the software, and that the software produced the speech “instantaneously.”
Carlo Calenda, leader of Lombardo’s Agione party, said on Twitter that the senator’s remarks on the floor were “impeccable” and added: “Are we seeing progress or a step back? It’s not clear to me yet,” he added.
Political cartoons about world leaders
Lombardo said he was impressed by recent developments in AI, but called for a deeper understanding of the technology to strike the right balance between regulation and innovation.
Italy has been at the center of global regulatory efforts on AI in recent months, with the country’s data protection authority temporarily banning ChatGPT and launching an investigation into the application for alleged breaches of privacy rules.
On Wednesday, the European Union’s technology chief Margrethe Vestager said a draft voluntary code of conduct for generative AI would be drafted “within the next few weeks”, with a final proposal for industry to sign “very soon”. said it was possible.
(Reporting by Federico Maccioni, Editing by Alvise Armellini and David Holmes)
Copyright 2023 Thomson Reuters.
