If there’s a single thread running through this week’s episodes, it’s the discipline required to turn AI ambition into durable value—and the growing consensus that speed, on its own, is not a strategy.
Daniella Harkins of LiveRamp makes the case that there is no such thing as a perfect AI strategy yet, and that the leaders who win are the ones who anchor to real business problems and a solid data foundation rather than chasing the loudest vendor claims. Forrester’s Gina Bhawalkar extends that argument into the design studio, warning that organizations optimizing for the velocity of AI risk becoming extraordinarily efficient at producing forgettable experiences unless human judgment stays at the center. And Forrester’s Chuck Gahun pushes the question further still, asking what happens when the “customer” evaluating your brand isn’t a person at all but an AI agent working on their behalf—and why that shift rewards clarity and strategy over urgency. For the bonus pick, The Artificial Intelligence Show captures the same tension at the industry level, as a government-gated model release and new research on autonomous agents show AI colliding with operational and institutional reality.
LiveRamp’s Daniella Harkins on distinguishing AI hype and real innovation
Daniella Harkins, SVP of Product Go-To-Market at LiveRamp, joins Greg to cut through the noise of a market where every platform claims to be a revolutionary AI silver bullet. Her central provocation is a liberating one: if you think you have a flawless, multi-year AI strategy today, you’re fooling yourself, because the ground is shifting too quickly for that kind of certainty. Rather than getting paralyzed by that reality, she argues, leaders should resist jumping straight to the technology and instead start from the business challenges they’re actually trying to solve, picking one or two high-value use cases to test into and leaning heavily on partners along the way.
Underneath the strategy talk sits a hard prerequisite Harkins keeps returning to: without the right data foundation, AI simply helps you make the wrong decisions faster. She pushes back on the increasingly common notion that identity no longer matters, calling it a false narrative and arguing that stitching data to a durable identity strategy matters more now than ever—both for marketing that delivers a return and for the consumer trust that underpins it. The conversation closes on what all of this does to the marketer’s role, as Harkins describes AI as the thing that finally frees senior leaders from the tactical grind and gives them room to move into genuine strategy, provided they’re willing to give their teams specific, confidence-building assignments rather than a vague mandate to “go use AI.”
Forrester’s Gina Bhawalkar on building great CX with an AI-enabled workflow
Recorded at Forrester CX Forum East in Brooklyn, this conversation with Gina Bhawalkar, Principal Analyst at Forrester, opens with a pointed question: as generative and agentic AI accelerate everything, are brands at risk of becoming incredibly efficient at producing experiences customers won’t remember? Bhawalkar’s research suggests the danger is real, pointing to Forrester data showing fewer than one in five consumers trust AI-powered interactions like chatbots and virtual assistants, and warning that prioritizing speed over quality tends to produce experiences that miss real customer needs, fail accessibility standards, and erode already-fragile trust.
Her prescription is a human-led decision model in which AI acts as an accelerator, not a replacement for judgment. She shares a principle from a chief design officer at National Australia Bank—”never delegate understanding”—as a guardrail against letting AI hollow out the customer knowledge that should live in a team’s heads. Much of the practical advice centers on the design system as the critical, machine-readable foundation for scaling AI-generated experiences without losing brand distinctiveness, alongside a recommendation to map internal workflows and point AI at specific bottlenecks rather than everything at once. Bhawalkar also reframes how teams should talk about ROI, urging leaders to describe efficiency gains as effort reclaimed and reinvested rather than hours or headcount saved, and she rounds out the discussion with responsible-design tools like “bad headlines” workshops and an added “impact” dimension to the classic desirability-viability-feasibility framework. The throughline: your humans are still your superpower.
Forrester’s Chuck Gahun on AI agents as decision makers in the buyer’s journey
Chuck Gahun, Principal Analyst at Forrester, tackles what may be the most disorienting shift of all: what it means to market not just with AI, but to AI. He frames the AI agent as a fundamentally new kind of consumer—one driven by logic rather than emotion, shaped by value rather than identity, and evaluating brands through non-deterministic experiences it assembles from published content and third-party sentiment. That duality between human and machine buyers is pushing brands toward dual optimization, Gahun explains, building emotionally rich storytelling for people while creating separate, semantically clear experiences that agents can actually parse.
From there the conversation gets refreshingly concrete. Gahun lays out the three things his research found agents are looking for—context, fluency, and trust—and explains why the creative flourishes brands love (the “eco-friendly” descriptors, the clever product names) have to be standardized into the same language everyone else in the sector uses so machines can reconcile them. He walks through real examples, from a brand engineering FAQs to overturn negative sentiment in answer-engine results to General Motors running a cross-functional, dealer-powered optimization strategy, and he reframes this whole effort as a content-strategy business case rather than a tech-implementation project. Looking ahead to agentic commerce, Gahun is candid about the friction still in the way—consumer trust, payment protocols, and questions of liability when an agent-driven purchase goes wrong—and he lands on a philosophy Forrester is coaching leaders to adopt: strategy over urgency, choosing deliberately which products to sell on which channels rather than racing to turn everything on at once.
Bonus Pick: [The AI Show Episode 222]: GPT-5.6, Government Staggers AI Model Releases, Agents Are Transforming Work & Growing Data Center Backlash — The Artificial Intelligence Show
Hosts Paul Roetzer and Mike Kaput of Marketing AI Institute and SmarterX open this week on a release that matters as much for how it happened as for what shipped: OpenAI previewed its next-generation GPT-5.6 family, but rolled it out first to a small set of government-approved organizations under a managed access list. Roetzer and Kaput unpack what they see as the “soft nationalization” of AI and the increasingly political nature of the decision to release a model at all, weaving in Dean Ball’s essay on what a workable safety standard would actually require and the open question of what happens to competing Chinese models if U.S. releases keep slowing down.
The episode pairs naturally with this week’s Agile Brand conversations because it turns from the politics of frontier models to the operational reality on the ground—including new OpenAI Codex research on how AI agents are genuinely changing the shape of real work, and a data-center backlash now uniting Americans across the political spectrum. Where Harkins, Bhawalkar, and Gahun each argue for grounding AI ambition in discipline and durable foundations, Roetzer and Kaput show the same collision playing out at the level of the industry itself, as the technology runs headlong into institutional, infrastructural, and human constraints.
Taken together, these conversations describe a market that is growing up. The reflex to move fast simply because AI makes it possible is giving way to harder, more useful questions: Is the data foundation sound? Is a human still making the decisions that matter? Have we actually decided who—or what—we’re building for? The leaders who thrive won’t be the ones with the flashiest tooling, but the ones who pair genuine innovation with the strategy and judgment to use it well. See you next week!
