AI safety fault lines at the AI summit in Dublin

Summary
This Reuters news dispatch, reported by Carmel Crimmins at the International AI Summit in Dublin, outlines conflicting viewpoints among tech leaders and European regulators regarding AI safety and regulation speed. The report highlights commentary from Microsoft Vice Chair and President Brad Smith, EU tech chief Henna Virkkunen, Google DeepMind Co-Founder and Chair Demis Hassabis, and OpenAI CFO Sarah Friar.

What is shown

• [00:00] Brad Smith addressing the summit stage regarding AI safety rules.
• [00:05] Host Carmel Crimmins standing on the summit exhibition floor in Dublin, introducing the divide on AI governance.
• [00:20] EU tech chief Henna Virkkunen delivering a speech on the EU AI Act and risk mitigation.
• [00:44] Sir Demis Hassabis speaking in a seated summit panel proposing an adaptable standards body.
• [01:11] OpenAI CFO Sarah Friar participating in a fireside chat discussing pacing versus falling behind malicious actors.

Claims & numbers

• Brad Smith asserts there are currently more safety regulations governing a carton of milk than AI systems [00:00].
• Henna Virkkunen claims the EU’s existing legal framework, the AI Act, is equipped to tackle rogue agents by regulating high-risk systems and mandating providers assess and mitigate systemic risks from advanced models [00:20].
• Demis Hassabis argues that legislation enacted three years prior becomes out of date due to the rapid speed of AI development, advocating instead for a fast-adapting standards framework [00:44].
• Sarah Friar claims that excessively slowing deployment risks creating an imbalance where defensive capabilities lag behind malicious actors, particularly in cyber contexts [01:11].

Notable quotes

• [00:00] “There are more safety rules for a carton of milk than for AI.” — Brad Smith
• [00:44] “If you legislate for something three years ago, it will be out of date today and actually probably won’t even address what the real problems are.” — Demis Hassabis
• [01:11] “We need to make sure that doesn’t slow us down to the extent that we then end up behind the folks who are on the malicious side of deployment.” — Sarah Friar

Assessment
This is an authentic news segment from Reuters covering policy debates at an international tech conference. It captures edited excerpts from official stage panels and speeches rather than live technical demonstrations or benchmark tests.

Described by gemini-3.8-flash on 2026-10-15 from the video’s audio and frames.