AI in Media Buying: A Double-Edged Sword for Agencies?
Digiday+ members, get ready for your weekly dose of media buying insights! This week, we delve into the fascinating yet complex world of AI in programmatic advertising, as discussed at the recent Programmatic Marketing Summit in New Orleans. But here's where it gets controversial... While AI promises to revolutionize the industry, it also raises concerns about transparency, accountability, and the future of human expertise.
The AI Tightrope: Promise or Threat?
The summit, held under Chatham House rules to encourage open dialogue, revealed a media agency world grappling with the implications of generative AI. And this is the part most people miss... While machine learning is already deeply integrated into buying processes across platforms like Meta and Google, the emergence of generative AI and agentic technologies is causing both excitement and unease.
Where AI Fits (and Doesn’t Fit) in Programmatic
Agency executives largely agree that AI agents should steer clear of the actual transaction point between media buyers and sellers. Instead, they see value in using AI for pre- and post-transaction tasks, such as campaign planning and performance data analysis. However, even this level of involvement requires significant training for agency employees—not just on AI tools but also on the fundamentals of programmatic workflows to ensure effective human oversight.
The Smoke Screen Effect
Bold statement alert: AI is increasingly being used as a cover for opacity in pricing and performance optimizations. Agencies express concern that platforms like The Trade Desk are leveraging AI to obscure processes, leaving them with less control and fewer ways to challenge decisions. Thought-provoking question: Is AI becoming a tool for obfuscation rather than transparency? One executive highlighted the challenge of holding AI accountable for optimizations, as it lacks the ability to explain its reasoning—unlike a human, who can be asked, “Why did you make that decision?”
AI’s Limitations: Probabilistic, Not Predictive
AI agents excel at fetching and gathering data but fall short when it comes to critical thinking. As one participant put it, “Agents don’t think—they’re probabilistic.” This means they’re not yet reliable for handling novel or complex tasks. Controversial interpretation: Are we overestimating AI’s capabilities by treating it as a one-size-fits-all solution? Another concern is AI’s tendency to “hallucinate,” especially when working with outdated or unreliable data. This flaw could lead to costly mistakes, particularly in high-stakes areas like legal compliance and financial planning.
The Future of AI in Media Planning
While some programmatic partners offer tools to automate entire media plans, the industry is still far from a one-click solution. But here’s the kicker: Legal and financial teams remain wary of AI’s reliability, and for good reason. As one executive noted, “All it takes is for the AI to hallucinate one zero, and you’re in big trouble.” Counterpoint: Could AI’s limitations actually be a blessing in disguise, forcing agencies to maintain a human touch in strategic decision-making?
Training AI: Low and Slow
To train AI effectively, agencies are adopting a “low and slow” approach, using older, cleaner data sets to minimize errors. Analogies to the rescue: One participant compared it to “good barbecue—low and slow,” emphasizing the importance of quality over speed. Question for the audience: How can agencies strike the right balance between leveraging AI’s efficiency and preserving human expertise?
Accountability in the Age of AI
A recurring theme was the need for accountability. Managers must ensure that employees don’t use AI as an excuse for mistakes. Bold statement: “If we allow everyone to use that as an excuse, we’re going to see a ton of errors.” Controversial question: Are we risking the erosion of foundational knowledge among younger media planners who rely too heavily on AI?
Industry Moves and Insights
Shifting gears, let’s look at recent industry developments:
- Omnicom’s acquisition of IPG led to 4,000 layoffs and the closure of IPG’s Magna unit, sparking public backlash.
- Havas acquired U.K. experiential agency Bearded Kitten and French data firm Unnest to bolster its tech and data capabilities.
- Personnel updates: Dentsu appointed Kara Osborne Gladwell as Global Product Architect Officer and brought back Tia Castagno as Global Innovation President. Monks hired Thiago Correa as SVP of Media for EMEA, and Influencer appointed Ryan Fitzpatrick as CFO.
Direct Quote to Ponder
“This merger is a reminder that scale doesn’t fix fragmentation. AI only works when the underlying customer data is unified, governed, and understood. The organizations that win in the next phase of marketing won’t be the ones with the most tools—they’ll be the ones with the clearest picture of their customers.” – Tony Owens, CEO of Amperity
Quick Takes
- The Trade Desk is easing its grip on pricing due to pressure from rival DSPs.
- Criteo is rebranding itself as an AI-powered commerce media platform.
- The creator economy is set to expand significantly by 2026, according to Krystal Scanlon’s analysis.
Final Thought
As AI continues to reshape the media buying landscape, the industry must navigate its promises and pitfalls. Question for you: How can agencies harness AI’s potential while safeguarding transparency, accountability, and human expertise? Share your thoughts in the comments—let’s spark a debate!
