Hollywood ‘fuels the fire of fear’ about artificial intelligence, peers hear

Hollywood has fanned the flames about the dangers of artificial intelligence (AI) in the minds of a generation of “engineers, computer scientists and super geeks”, ministers heard.

Film depictions of artificial intelligence such as The Terminator help solidify “hopes and fears about the impact artificial intelligence may have on us” as it considers plans to regulate the emerging technology, the House of Lords has been told.

The Lords of Parliament are being urged to support Lord Richmond Holmes’ proposal to create a new regulator, the Artificial Intelligence Authority.

The Tory peer’s Artificial Intelligence (Regulation) Bill, which begins in Parliament on Friday, calls for powers to advance the regulation of artificial intelligence in the UK and to assess and monitor potential risks to the economy.

Some peers have expressed concerns about generative AI, including the need to ensure artists whose work is used as prompts are paid fairly and that the technology should be prevented from exploiting images of child sexual abuse.

But Northwood’s Conservative peer Lord Rangel said the technology’s proponents now needed space to innovate.

He told the House of Lords: “If you, like me, come from a certain generation, the seeds of these fears and fascination with the power of artificial intelligence have long been sown by numerous Hollywood films that captured our imagination of what artificial intelligence might do to us. hopes and fears.”

He cited HAL 9000’s “disturbing surrender” in “2001: A Space Odyssey” and the “ultimate hellish future in which machine intelligence takes over the world in the form of Skynet” in the “Terminator” movies.

Lord Ranger added: “These and many other futuristic interpretations of artificial intelligence have helped stir the minds of engineers, computer scientists and super-geeks, many of whom have created the world’s largest technology companies.”

While he said he supported the aims of the bill and that longer-term regulatory guidance may be needed, Lord Rangel said he did not think it would be possible to regulate AI through a single agency.

He also criticized the labeling system it will introduce to ensure that anyone involved in AI training must provide authorities with records of all third-party data and intellectual property (IP) they use and provide assurances that its use has been Obtain consent.

“This will not… help us work with industry and trade bodies to build trust and confidence in technology,” the Conservative MP said.

Other peers also backed the bill, with crossbench Lord Freyberg telling the House of Lords: “It stands to reason that if artists’ IP (intellectual property) is used to train these models, it is fair that they should be compensated.” Be credited and provide the option to opt out.”

Crossbench peer Baroness Kidron, meanwhile, said she wanted “more clarity that criminal material such as creating viruses, CSAM (child sexual abuse material) or inciting violence, whether created by AI or not, is a crime .”

The filmmaker and children’s rights activist cited a report from the Stanford Internet Observatory that identified “hundreds of known sexual images of children in an open dataset used to train popular artificial intelligence text-to-text models.” Images of Abuse Material”.

She added: “Reports indicate that there was a strong possibility of removing such images, but they didn’t bother. Now these images are proliferating on a massive scale. We need to put some rules in place for the development of artificial intelligence.”

Lord Holmes, the bill’s sponsor, compared the emergence of artificial intelligence to the emergence of steam power during the Industrial Revolution and urged peers to back his proposals.

He said: “If artificial intelligence is to human intelligence what steam is to human strength, you get it. Steam has literally changed time. It’s time for us to act, which is why I’m bringing this bill to the House today. House.”

The government believes the non-statutory approach to AI regulation provides “critical adaptability” but has pledged to review it.

A government spokesman said: “In line with standard procedure, the government’s position on this bill will be confirmed during the debate.”