- By Siona McCallum, Chris Valance
- technology reporter
image source, Getty Images
OpenAI CEO Sam Altman
The head of the company that runs ChatGPT said it could consider leaving the EU if it fails to comply with planned artificial intelligence (AI) legislation.
It could also require generative AI companies to reveal what copyrighted material was used to train systems to create text and images.
“The current draft EU AI law will be over-regulatory,” said Sam Altman of OpenAI, Reuters reported.
“But I hear it will be withdrawn.”
Many in the creative industry have accused AI companies of using the work of artists, musicians and actors to train systems to imitate their work.
However, according to Time magazine, Altman is concerned that OpenAI is technically incapable of complying with some of the AI Act’s safety and transparency requirements.
image source, Upcoming Publications/Getty Images
Demonstrators outside UCL spoken by Sam Altman
Speaking at an event at University College London, Altman added he was optimistic that AI could create more jobs and reduce inequality.
He also met with Prime Minister Rishi Sunak and the heads of AI companies DeepMind and Anthropic to discuss technology risks, from disinformation to national security and even “existential threats,” and how to manage them. Discussed the voluntary actions and regulations necessary for
But Sunak said AI could “positively transform humanity” and “create new opportunities in a range of areas to improve public services and deliver better outcomes for the British public”.
image source, Downing Street No 10
AI bosses met Prime Minister at 10th place
At the G7 summit in Hiroshima, leaders of the US, UK, Germany, France, Italy, Japan and Canada agreed that developing “trustworthy” AI must be an “international effort.”
And the European Commission is aiming to reach an AI deal with Google’s parent company, Alphabet, before EU law comes into force.
EU industry chief Thierry Breton, who met with Google CEO Sundar Pichai in Brussels, said international cooperation is essential in regulating AI.
“Mr Sander and I agree that we cannot wait until AI regulations are actually applied and we are working with all AI developers to voluntarily conclude AI agreements ahead of legal deadlines. We agreed to formulate it,” Bretton said.
Tim O’Reilly, a Silicon Valley veteran, author, and founder of O’Reilly Media, said building a regulatory body that mandates transparency and ensures accountability would be the best start.
“AI’s scare-mongering practices, coupled with regulatory complexity, can lead to analysis paralysis,” he said.
“Companies developing advanced AI will work together to create a comprehensive set of metrics that can be reported to regulators and the public on a regular and consistent basis, and to update those metrics as new best practices emerge. We have to develop a process.”
