By Philip Blenkinsop
LULEA, Sweden (Reuters) – The United States and European Union should push the artificial intelligence (AI) industry to adopt a voluntary code of conduct within months to provide safeguards while new laws are developed, EU tech chief Margrethe Vestager said on Wednesday.
The European Union’s AI Act, with rules on facial recognition and biometric surveillance, could be the world’s first comprehensive legislation governing the technology, but is still going through the legislative process.
“In the best of cases it will take effect in two and a half to three years time. That is obviously way too late,” Vestager told reporters before a meeting of the joint EU-U.S Trade and Technology Council in Sweden. “We need to act now.”
EU industry chief Thierry Breton said last week that Alphabet and the European Commission aimed to develop an AI pact as concerns mount about the impact on society particularly from generative AI, like ChapGPT, that create content.
Leaders of the G7 nations called earlier this month for the development of technical standards to keep AI “trustworthy”, urging international discussions on topics such as governance, copyrights, transparency and the threat of disinformation
Vestager said there needed to be agreement on specifics, not just general statements, suggesting the European Union and the United States could help drive the process.
“If the two of us take the lead with close friends, I think we can push something that will make us all much more comfortable with the fact that generative AI is now in the world and is developing at amazing speeds,” she said.
Vestager, a European Commission vice president, said a code of conduct come emerge quickly while governments and legislators from the EU to Canada to India establish rules.
“That is the kind of speed you need, to discuss in the coming weeks, a few months, and of course also involve industry … in order for society to trust what is ongoing,” she said.
(Reporting by Philip Blenkinsop; Editing by Mark Potter)
