computer
Sunak Advisor: Artificial Intelligence Could Kill People
“There are many different risks with AI.” photo
© Sebastian Gollnow/dpa
Matt Clifford calls for global regulation of artificial intelligence. Among other things, he warns of new instructions for bioweapons, but also sees opportunities in certain areas.
The technology advisor to British Prime Minister Rishi Sunak warned in no uncertain terms of the deadly dangers posed by artificial intelligence. The technology could create “dangerous threats that could kill a lot of people,” Matt Clifford told UK broadcaster Talk TV.
Artificial intelligence must be regulated globally as soon as possible, otherwise “very powerful” systems could emerge, he warned. It is currently not clear how AI models can be controlled.
“You can already use AI to develop new instructions for bioweapons or to launch large-scale cyber attacks,” said Clifford. “There are many different risks with AI. In the industry, we often talk about short-term and long-term risks, and the short-term risks are already pretty scary.” Asked about the risk of AI wiping out humanity, Clifford said, “I don’t think it’s zero.”
“Properly applied, artificial intelligence is extremely useful”
Prime Minister Sunak also wants to talk to US President Joe Biden about artificial intelligence during an upcoming visit to Washington. Leading experts recently warned that artificial intelligence is a potential threat to humanity and must be given a global priority alongside other risks of societal proportions such as pandemics and nuclear war. However, an expert from the University of Oxford described this warning as a “PR campaign”.
The chatbot ChatGPT, which can formulate sentences at human level, had triggered a new hype about artificial intelligence in the past few months.
When used correctly, artificial intelligence is extremely useful, emphasized consultant Clifford. “It is conceivable that AI will cure diseases, boost economic production and help us develop a carbon-neutral economy,” he said.