The Brand & AI Safety Summit returns to London on September 17, 2026, at Sea Containers on the South Bank, expanding its scope to tackle the collision of brand safety and artificial intelligence in digital advertising. The full-day event brings together senior marketers from Unilever, Mondelēz, Haleon, Google, Meta, and TikTok to address what organizers call a widening gap between how media is bought and how its value is actually measured.
This year's summit reflects a structural shift. Content tracks now run by industry vertical - healthcare, commerce, sports, financial services, and creators - matching how agencies and brands organize their spending. A parallel track of small-group boardroom sessions, capped at thirty seats each, will run alongside the main stage.
The measurement problem marketers can't ignore
The opening fireside chat sets a blunt tone. Lucia Mastromauro, former VP of Advertising at Deliveroo, and summit president Rob Rasko will argue that brand safety KPIs have quietly replaced performance math. "Suitability scores and adjacency ratings were built to answer one question, did anything embarrassing happen, not whether the money actually returned value," the session description states. Platforms, they contend, have let safety reporting stand in for proof that budgets reached something worth the price.
That tension runs through multiple sessions. A Pixis presentation at 2:25 PM tackles the explainability gap in AI decisioning. Daniel Fitzgerald, Director of Strategic Partnerships for EMEA, said agencies increasingly field client questions about why algorithms made specific choices - without having full answers themselves. "The real tension is the explainability gap it creates, and whose job it is to close it," the session brief reads.
Creators, healthcare, and the real edges of AI
A morning panel on creators features TikTok's Imrati Anand, Responsible Marketing Advisory's Alice Brady, and Publicis Groupe's Bailey Chatwin. They will explore how AI tools are reshaping content creation and what authenticity means when the creative toolkit expands. The session frames AI not as a threat to human creativity but as something that "could make human creativity bigger, not smaller, helping more people bring more ideas to life."
Healthcare gets dedicated attention. A closed-door morning session with Haleon's Silas Lewis-Melius and CMI Media Group's Theo Strikos will surface war stories about how market decisions actually get made in regulated categories. A later main-stage panel asks what it takes to ship bold, culturally resonant work when internal risk tolerance, not creative talent, is the real constraint. Novartis' Cannes Lions Pharma Grand Prix-winning campaign serves as the working case study.
Channel Factory presents a session arguing that the AI revolution is fundamentally a text revolution - LLMs run on language, not pixels - and that Reddit represents the internet's richest source of authentic human text that brands have been too cautious to tap. Meanwhile, Google's Parvathy Bavishi joins a panel on AI slop, examining what happens when speed and volume optimize for feed velocity instead of craft.
Sports and commerce: beyond the media buy
An afternoon sports panel with Omnicom Media's Saiful Ahmed and Jellyfish's Natasha Wallace argues that most brands still treat sports like a media buy instead of a relationship. The session description is characteristically direct: "The cynical case: most brands treat sports like a media buy instead of a relationship, and wonder later why it didn't land."
Commerce media gets its own private lunch session and a main-stage panel. Mondelēz International's Jennifer Ballou joins Ocado Ads' Ben O'Mahoney and Grace & Co's Darren Jacobs to examine what it takes to build and scale a retail or DTC story. The summit closes with Hakim Charles, founder of regulatory technology startup Vellum Intelligence, and Rasko delivering a closing vision on where brand and AI safety head next.
Why this matters for marketing and communications professionals
The summit's programming reflects a real operational problem: brand safety tools were built to prevent embarrassment, not to prove return on investment, yet they are increasingly used as proxy KPIs. For marketing and communications leaders, the practical takeaway is that the industry's measurement infrastructure needs rebuilding. Sessions on AI for Marketing and explainable AI decisioning point toward tools that give agencies their own layer of visibility when platforms keep optimization logic behind closed doors. The networking mixer running from 4:30 to 7:00 PM at the 12th Knot offers direct access to speakers who have led media at Unilever, Deliveroo, and Haleon - people who have already navigated the shift from safety theater to performance accountability.
Your membership also unlocks:
