Advertising Automation Enters Agentic Era as AI Platforms Target Full Campaign Control

A new generation of artificial intelligence platforms is moving beyond automated bidding and placement to assume end-to-end control of advertising campaigns, marking a fundamental shift in how brands reach consumers and raising questions about the role of human strategists in media planning.

Seedtag last week launched an agentic AI platform called Liz that uses what the company describes as "neuro-contextual intelligence" to plan and activate media campaigns through conversational text and voice interfaces. The system attempts to identify consumer emotions at the moment of purchase consideration, combining machine learning and natural language understanding to predict human interest, emotions and intent from content signals. "All the signals are combined to bring the signals together," said a Seedtag executive, adding that "the data is fine-tuned through models."

The development follows years of advertiser criticism over Google's aggressive push toward automation, which many marketers argued stripped away strategic control. Now, generative AI is enabling companies to automate what Seedtag and others call the "entire ad supply chain," moving from tactical execution to strategic planning.

Meanwhile, companies deploying AI at scale are confronting a new operational reality: the cost of AI usage itself. Automation platform maker Zapier has introduced internal dashboards tracking how many tokens—AI's unit of computational measurement—employees consume, signaling that widespread AI adoption brings its own budgetary pressures. The metric reflects a broader industry shift as firms that once focused on AI training costs now grapple with inference expenses at scale.

(The advertising industry has experimented with contextual targeting and programmatic automation for over a decade, but agentic AI represents a qualitative leap by combining planning, activation and optimization in a single autonomous system. Seedtag's platform relies on what it calls a "contextual graph" that provides network-level analysis in near real-time.)

The push toward autonomous advertising comes as AI applications expand across adjacent domains. L'Oréal and Nvidia recently broadened their partnership beyond digital marketing to include AI-driven computational chemistry for molecular discovery, while Nvidia announced collaborations with fashion industry players on virtual try-on technology using physics-based modeling. Separately, AI-enhanced radar systems are being developed to improve perception for autonomous vehicles and robotics, illustrating how reasoning models are being applied to physical-world sensing challenges that have long resisted traditional signal processing approaches.

The advertising automation wave also intersects with broader enterprise AI adoption patterns. As companies integrate AI tools across workflows, they face dual pressures: demonstrating productivity gains while managing the infrastructure costs and governance challenges that come with widespread deployment. The emergence of token-tracking dashboards at firms like Zapier suggests that AI's operational overhead may temper some of the cost-reduction promises that initially drove adoption.