Experts respond to launch of EU AI Act, discuss impact on businesses

The European Artificial Intelligence (AI) Act officially came into force yesterday, marking a significant regulatory development as the EU becomes one of the first regions worldwide to introduce comprehensive AI regulation.

The new legislation, designed to provide a structured framework for AI system governance, is set to impact businesses across various sectors operating within the EU.

Commenting on the new legislation, 3M's Global Digital Automation & Innovation Senior Manager, Paul Cardno, stressed the need for clear guidance.

"With nearly 80% of UK adults now believing AI needs to be heavily regulated, the introduction of the EU’s AI Act is something that businesses have been long-waiting for," Cardno told TechDay.

He highlighted the Act's potential to foster confidence in the re-imagination of business processes and the breaking away from entrenched structures.

While acknowledging imperfections in the Act, Cardno stressed the importance of a clear framework to ensure AI has a safe and positive influence across organisations.

Head of AI Labs at Pegasystems and assistant professor of AI at Leiden University, Peter van der Putten, discussed the different risk categories established by the Act - unacceptable, high, intermediate, and low-level risk applications. "It is essential for organisations to pay close attention to how this new regulation will impact them," he said.

Van der Putten noted that while few uses of AI are outright prohibited, businesses must focus on high-risk categories, necessitating tighter controls, governance, and documentation for accountability, oversight, accuracy, and bias testing.

Jonathan Armstrong, Partner at Punter Southall Law and a specialist in compliance and technology, pointed out the EU AI Act's complex legislative nature.

Armstrong equated the Act's structure to previous EU regulations such as product safety, competition, and GDPR. He acknowledged potential challenges, including under-resourced regulators and patchy enforcement, but also credited the Act for shifting perceptions around AI risks and responsibilities. Armstrong advised businesses to start preparing now by assessing their current AI use, conducting compliance gap analyses, and creating bespoke Action Plans to ensure readiness.

The financial sector, in particular, stands to be significantly impacted by the new regulation.

Bob Stark, Global Head of Market Strategy at Kyriba, shared insights specific to corporate finance. Stark noted a substantial increase in CFOs investing in AI—51% this year compared to just 15% in August 2023—indicating a growing recognition of AI's importance in financial process automation and data-driven decision-making.

He assured finance teams that their use of AI, predominantly for improving human decision-making or reviewing AI model results, is not considered high risk under the new law.

Speaking to TechDay, Stark added he believes the Act will "prompt more transparency" from General Purpose AI (GPAI) providers about how AI models are trained and the data used - providing further validation for AI adoption in corporate finance.