At Cannes Lions 2026, Mother London picked up the Film Grand Prix. Twice. Both wins were for the same client: Anthropic's Claude. The two spots, "Can I Get a Six Pack Quickly?" and "How Can I Communicate Better with My Mom?", both ran during the Super Bowl, and both grew out of a single creative platform built on a single business decision: Anthropic's pledge not to run ads inside Claude.
Film Jury President Pelle Sjoenell described it as "an ad about where ads shouldn't exist". It was corporate aikido, using a competitor's own announcement as the weapon. The execution was precise, the risk was genuine, and the work ended up reshuffling how the most consequential technology category of our age is perceived.
That's a remarkable outcome for what is, at its core, a values statement.
The decision that made the work possible
In February 2026, Anthropic published a blog post titled "Claude is a Space to Think." The commitment was straightforward: users won't see sponsored links next to their conversations, and Claude's responses won't be shaped by advertiser relationships or include third-party product placements they didn't ask for.
The timing was deliberate. OpenAI had just confirmed it would introduce advertising to ChatGPT for free and Go-tier subscribers. Anthropic used the Super Bowl, where OpenAI was also running a spot, to make the contrast unmissable.
What Anthropic wrote in that blog post is worth reading carefully: "The history of ad-supported products suggests that advertising incentives, once introduced, tend to expand over time as they become integrated into revenue targets and product development, blurring boundaries that were once more clear-cut. We've chosen not to introduce these dynamics into Claude."
That's not a product feature. It's a position. And it's the kind of position that creative agencies should be able to articulate about themselves, but mostly can't.
Why capability was never going to be the differentiator
The reason Anthropic's values-led positioning works in this category is structural: AI capability is commoditising fast.
Every major model is built on roughly the same transformer architecture, increasingly fine-tuned on similar data, priced on similar consumption models. Sam Altman himself drew the comparison to electricity at the BlackRock US Infrastructure Summit in March 2026: AI should eventually be priced like a utility, by consumption, not by flat fee. RBC's analysis puts it more starkly: "the value isn't captured primarily by the producer, but by the systems built on top of it."
This is the exact dynamic we saw with electricity itself. The 20th century's most transformative technology became infrastructure, invisible, unremarkable, billed monthly. Nobody built a lasting brand on "we have electricity." They built brands on what they did with it.
So if you're a company in the AI space, or a network racing to build AI capability, differentiation on the technology was always going to be a short-term play. Anthropic seems to have understood that earlier than most.
What the holdcos are actually saying about themselves
The week after Cannes, I went back and read the values and positioning statements of the major networks. What I found was that, for all the stick the legacy holdcos take for being slower to move on AI, most of them have spent decades building genuine identity, not just capability.
WPP positions around being open, optimistic and extraordinary — a creative platform that's remained consistent through significant business transformation. Publicis Groupe has long led with "Viva la Différence" — entrepreneurship, diversity, generosity, excellence and ethics, a values architecture that pre-dates the AI conversation by years. Omnicom's stated values — client-centricity, creativity, collaboration, accountability, integrity, inclusion — are the kind of list that can feel like boilerplate, but they've held stable across acquisition cycles and market shifts. Havas frames its purpose as "Growth, Powered by Desire", with growth explicitly built on social and environmental responsibility. And Dentsu leads with what amounts to a purpose statement rather than a tech statement: "an invitation to the never before."
None of these are AI statements. That's the point. They're durable identity positions that will still make sense in five years, whatever the model landscape looks like.
The Brandtech question
The contrast worth examining is The Brandtech Group, which markets itself as the world's #1 Gen AI marketing company and whose stated mission centres on helping brands do marketing "better, faster, cheaper" through technology.
That's a genuinely interesting positioning choice. Brandtech has been consistent and early on AI, and there are signs of values-layer thinking emerging underneath the capability pitch: they've launched an Ethical Gen AI package including a "Bias Breaker" tool, run AI for Good scholarship programmes, and positioned responsible use as part of their offer. Their founder David Jones has made the case publicly that 2025 would be "the year AI will upend advertising", and he's largely been right.
But the top-line identity remains anchored in speed and efficiency. The question I'd put to any network using "AI-first" as its primary identity is: when the capability is genuinely commoditised — when every network runs the same models at the same speed at the same price — what's left? Is there a values layer waiting to be surfaced? Because if the answer is no, "better, faster, cheaper" is a race to the bottom that gets harder to win as the market matures.
I don't know the answer for Brandtech specifically. But it's the right question to be asking of any network whose B2B positioning leans hardest into the technology itself.
The note to take
Anthropic didn't win at Cannes because Claude is the smartest model in the room. It won because it stood for something specific, sold that stand rather than the spec sheet, and found a creative team in Mother London willing to make work brave enough to deserve the position.
The irony is that agencies — businesses which help clients build exactly this kind of positioning every day — are among the worst at applying it to themselves. Networks that have spent decades telling CPG clients that differentiation comes from brand identity, not product parity, are now racing to prove their AI parity with each other.
The lesson from Anthropic isn't about AI at all. It's about what happens when a company identifies one thing it believes — "we won't interrupt you with commercial messages" — and commits to it publicly, commercially and creatively, before the category forces the conversation.
That kind of decision is available to any network. It requires working out what you actually stand for. Then being willing to stand there.