Elections in Britain and the US next year could be marked by AI-powered disinformation, experts have warned, as deeply fake images, text and videos spread at the behest of swarms of AI-powered campaign bots.
Sam Altman, CEO of OpenAI, creator of SatGBT, told a congressional hearing in Washington a few days ago that the models behind the latest generation of AI technology could manipulate users.
“The general ability of these models to manipulate and coerce, to interact with each other and provide misinformation, is of great concern,” he said.
The UK will lead the fight to reduce the risks of artificial intelligence, Prime Minister Rishi Sunak said on Thursday.
Professor Michael Wooldridge, director of the AI Research Foundation at the Alan Turing Institute in the UK, said his main concern was AI-powered misinformation.
“In terms of my concerns about AI, this is at the top of the list,” he said. “We have elections coming up in the UK and the US and we know that social media is an incredibly powerful channel for misinformation, but we now know that AI can generate information.” misleading on an industrial scale.”
Wooldridge says chatbots like ChatGPT can create personalized disinformation aimed at a conservative voter, a metropolitan labor voter or a Republican supporter in the Midwest.
“Coffee trailblazer. Social media ninja. Unapologetic web guru. Friendly music fan. Alcohol fanatic.”
More Stories
Wisely visits the British forces in the “Eastern sector”.
India and China top the 10 nationalities receiving work permits for foreigners in 2021.
The Guardian: A 2% tax on Britain’s rich would raise £20 billion a year