Stricter European Union regulations on artificial intelligence came into effect yesterday, mandating companies to ensure their AI systems are both safe and transparent, while also requiring governments to establish enforcement mechanisms.
The latest phase of the EU’s landmark Artificial Intelligence Act, first introduced last year, aims to protect consumers from potential harms posed by AI, while still encouraging responsible innovation. Among the new provisions is a rule requiring AI-generated voices—especially those designed to sound human—to be clearly identified as artificial.
“With the sophistication of AI, things like artificial voices can be so nuanced that they can convince you it’s human,” said John Mitchell, CEO of digital agency All human, which is currently developing an AI-powered energy comparison tool using an Irish-accented chatbot. “Organisations and service providers must now declare what is real and what is not,” he added.
The Act bans AI systems deemed to pose an unacceptable risk to safety, human rights, or livelihoods, while placing strict transparency and compliance obligations on both large tech firms and smaller businesses.
Maureen Daly, a partner at Pinsent Masons Ireland specialising in AI and intellectual property law, said the rules apply to companies of all sizes, including start-ups and SMEs. “They’re going to have to comply as well,” she noted, warning that non-compliance from August 2026 could result in fines of up to €15 million or 3% of a company’s global turnover.
In addition to private sector responsibilities, the new rules require EU member states to set up enforcement structures, including appointing regulatory bodies and legislating penalties for violations. However, concerns have been raised that Ireland is lagging behind in its preparations.
“There needs to be urgency from the Government,” said Dr. Kris Shrishak, senior fellow at the Irish Council for Civil Liberties. “It’s not just about stopping harmful AI; it’s also about supporting positive development through initiatives like regulatory sandboxes for businesses.”
The Department of Enterprise said Ireland is meeting its obligations and cited ongoing efforts including working groups, EU consultations, and resource planning.
While the EU tightens oversight, a contrasting approach is emerging in the US. President Donald Trump has pledged to reduce regulation in an effort to “win the AI race,” raising concerns about potential competitive disadvantages for Europe.
Caroline Dunlea, Chairperson of Digital Business Ireland, acknowledged the importance of regulation but urged caution. “A balanced attitude is essential to avoid overregulation that could stifle innovation,” she said.
Mr. Mitchell agreed, saying the EU’s current strategy strikes the right balance between progress and protection. “In the long term, that’s the way to go with AI—it’s very unregulated today.”
A feature report on the new AI rules will air on This Week on RTÉ Radio 1 at 1 p.m.
