SAN JUAN, PR — In a digital age where artificial intelligence (AI) is the buzzword on every tech enthusiast’s lips, it’s easy to get swept up in the futuristic allure of generative AI.
But according to Dave Simon, VP, Strategic Partnerships, Moloco, the true power player in advertising is not as much AI itself, but the machine learning technology that underpins it.
In this video interview with Beet.TV, Simon explains the difference, and why it matters.
The Core of AI in Advertising
“Really, the underlying technology of machine learning is what creates the value,” Simon emphasizes.
His viewpoint is grounded in the practical application of AI in advertising, where machine learning models leverage vast datasets to solve business problems. These algorithms can process information and make predictions at a scale and speed unattainable by human teams.
Simon explains, “What we’re doing is leveraging different data sets that are traditionally provided by marketers or by publishers and helping them solve a business problem by making faster predictions with much deeper insight and much bigger scale than a group of humans could do on their own.”
Doing the Data Dance
When it comes to training these sophisticated models, Simon is clear about the ingredients for success: “Our models rely on two things:
- “The first thing is high fidelity, high volume, high predictability data…
- “And then the other side of the equation is, did the prediction that we made leveraging that data come true?”
This dual-data approach is critical for closing the feedback loop and refining the machine learning process, ensuring that the models are continually improving and delivering more accurate predictions.
Transparency in the Age of AI
The topic of transparency, or the lack thereof, particularly in Google’s Performance Max product, has been a point of contention. “Google has had to now go back and open up more transparency with it,” says Simon. Despite these challenges, he acknowledges that platforms like Performance Max are designed to deliver profitable outcomes, simplifying the process for marketers by using machine learning to determine when and where ads should be served.
Simon points out a fundamental shift in how AI-driven platforms like Performance Max, Instagram, and YouTube operate. “They don’t preemptively determine who that audience is going to be. They let the machines make an association between the outcome you want and the user who’s going to receive an ad at that moment,” he states.
This move away from traditional cohort-based targeting is significant, emphasizing the efficiency of machine learning in real-time decision-making.
AI’s Role in the CTV Landscape
The application of AI in Connected TV (CTV) is still a developing story, with Simon noting, “There is no one person who has a perfect full view of everything happening with machine learning and AI across the industry because everybody’s doing something with it.” He sees two main use cases emerging:
- real-time ad opportunity prediction.
- the creation of better targeted cohorts by marketers and agencies using AI tools.
Looking ahead, Simon envisions a transformative role for CMOs and media buyers, one that transcends traditional marketing approaches. “They need to figure out how to unlock their data… to feed it into a machine learning model that I’m training to maximize the business value that I can create as a marketing effort,” he suggests.
You’re watching Beet.TV’s coverage of Beet Retreat San Juan 2024. This series and event is sponsored by Albertsons Media Collective, Moloco, OpenX and TransUnion. For more videos from the series, please visit this page.
