UK FCA greenlights AI sandbox to allow banks to work with partners like Nvidia

On Monday, the UK’s Financial Conduct Authority announced a partnership with U.S. chipmaker Nvidia to allow banks to experiment with artificial intelligence safely. The agency said it will establish a “supercharged sandbox” to give firms access to better data, technical expertise, and regulatory support to speed up innovation.

The FCA also stated that financial services institutions in the UK will be allowed to experiment with AI starting from October. The firms will use Nvidia’s accelerated computing and AI Enterprise Software products.

FCA offers solution to banks’ AI challenges

In their supercharged sandbox!

I guess other chip manufacturers aren’t allowed then… or did they just want to put Nvidia in there to show as though they know what they are talking about

— Wiseman (@thomwiseman) June 9, 2025

The financial regulator noted that the initiative is designed for companies in the discovery and experimental phase with AI. The FCA also acknowledged that a separate testing service exists for firms further along in AI development.

The agency said the new sandbox addresses a key issue for banks, which have faced challenges shipping advanced new AI tools to their customers amid concerns over risks around privacy and fraud. Large language models from companies like OpenAI and Google send data back to overseas facilities, which privacy regulators have flagged over how the information is stored and processed.

“This collaboration will help those who want to test AI ideas but who lack the capabilities to do so. We’ll help firms harness AI to benefit our markets and consumers while supporting economic growth.”

–Jessica Rusu, Chief Data, Intelligence and Information Officer at FCA. 

Nvidia’s CEO Jensen Huang is expected to give a keynote talk at a tech conference in London on Monday morning.

HSBC’s generative AI lead Edward Achtner told a London tech conference last year that he sees “a lot of success theater” in finance regarding artificial intelligence. 

Achtner hinted that some financial services firms are promoting advances in AI without tangible product innovations to show for it. He also mentioned that, while banks like HSBC have used AI for many years, new generative AI tools like OpenAI’s ChatGPT come with unique compliance risks.

Italy became the first country in the West to ban ChatGPT, citing a breach of Europe’s laws. The Italian Data Protection Watchdog ordered OpenAI to temporarily cease processing Italian users’ data amid investigations into a suspected breach of Europe’s strict privacy regulations.

UK proposes to regulate artificial intelligence

In huge news, the UK government was just defeated in the House of Lords for a second time over AI & copyright.

The government had removed amendments to the Data Bill that would require AI companies to disclose their training data. The Lords have now reinserted them.

Most… pic.twitter.com/cJE3ekUBSp

— Ed Newton-Rex (@ednewtonrex) May 12, 2025

UK’s technology secretary Peter Kyle announced last week that he intends to introduce an AI bill in the next parliamentary session to address concerns about issues including safety and copyright. Ministers had previously intended to publish a short bill that would have required companies to hand over large AI models such as ChatGPT for testing by the UK’s AI Security Institute. 

The bill addressed concerns that AI models could become so advanced that they pose a risk to humanity. The legislation was delayed, with ministers deciding to wait and align with Donald Trump’s administration in the U.S. because of concerns that any regulation might weaken the UK’s attractiveness to AI companies.

The UK government and the House of Lords are already in a standoff over copyright rules in a separate data bill. The legislation would permit AI firms to train their models using copyrighted material unless the rights holder opts out. On Wednesday, peers voted 221 to 116 to push for an amendment to force AI companies to be transparent about what material they use to train their models.

The government argued that the data bill is not the right vehicle for the copyright issue and has promised to publish an economic impact assessment and a series of technical reports on copyright and AI issues. A cross-bench peer, Beeban Kidron, said ministers have shafted the creative industries and decimated the UK’s second-biggest industrial sector.

Now, the bill faces the prospect of being shelved unless the Commons accepts the Kidron amendment or proposes an alternative.

The Lords minister for the digital economy and online safety, Maggie Jones, had urged peers to vote against the Kidron amendment after the government offered last-minute concessions in an attempt to prevent another defeat.

Kidron told peers it was their last chance to ask the government to provide a meaningful solution, and also urged ministers to establish concrete steps to subject AI companies to copyright rules.

Your crypto news deserves attention - KEY Difference Wire puts you on 250+ top sites