AI tools improve marketing workflows, but trust blocks wider adoption
Marketers are using AI for creative production and campaign planning at high rates, yet concerns about reliability and complexity are slowing adoption of the most autonomous AI systems, according to new research from Digiday+ based on a survey of 142 brand and agency professionals.
The findings reveal a gap between what AI can do and what marketers will trust it to do alone.
Where AI is already embedded
Generative AI has moved from experimental to routine. Eighty-two percent of respondents said their companies use generative AI for creative production, with 81% using it for marketing tasks and 75% for external communications.
Unilever's Beauty AI studio, built with marketing services group Brandtech, shows the practical impact. The system creates assets for paid social, programmatic display, and e-commerce. Selina Sykes, global VP and head of marketing transformation for Unilever's beauty division, said the tool shifted how teams work. "It's a different way of working. We used to send briefs off and get content back. Now it's this agile, iterative approach," she said. The studio now produces around 400 creative assets per product-up from 20 per campaign before.
Predictive AI adoption lags behind generative systems. Forty-eight percent of respondents use predictive AI for measurement and KPI analysis. Kroger Precision Marketing uses it to generate weekly email digests for suppliers that track performance across multiple metrics.
Media buying and planning remain underutilized areas. Only 35% of respondents use predictive AI for these tasks, and just 25% use generative AI. Some executives see potential, particularly for smaller teams. Adam Simon, former managing director at IPG Media Lab, said smaller brands and agencies could benefit more than larger ones, since they lack the resources to build custom automation that major platforms like Meta and Google already offer.
Agentic AI faces a trust problem
AI Agents & Automation systems that operate autonomously without human feedback are far less adopted. More than half of survey respondents-54%-said their companies don't use agentic AI at all.
The barrier is trust. Agentic AI runs tasks independently, which means errors compound. If an AI agent hallucinates-generating false information it treats as factual-every subsequent step breaks. Marc Maleh, global CTO at design agency Huge, said marketers must decide what data and actions to hand over to autonomous systems. "There is a governance concern, especially with agentic and data access," he said. "What do I want to give an agentic AI the rights to do on my behalf?"
Building trust takes time. Eric Lee, CTO at creative technology agency Left Field Labs, said teams start with smaller tasks and expand gradually. "Over the past year, we have seen more micromanaged agents, where you have an agent that can do a lot, but a layer of education and trust needs to be built before people are ready to say, 'Yes, go manage my inbox, or execute this code for me,'" he said.
Complexity is the second barrier. Agentic AI doesn't function as a single tool-it connects to multiple APIs, language models, and other systems. Sarah Mehler, CEO of Left Field Labs, said the technology is difficult to implement well. "Agentic AI is really hard to do, and not everybody can. There are very few examples of it actually working well in the world."
Some agencies have built working examples. Digital agency Monks recently partnered with Nvidia to create a 30-second film for Puma using agentic AI. Wesley ter Haar, co-founder and chief AI officer at Monks, said the process involved AI agents writing the script, creating mood boards, and directing photography-with agents coordinating across tasks.
What marketers need to prepare for
AI is shifting from assistant to operator. Matt Maher, founder of M7 Innovations, said brands need to prepare infrastructure now for systems like ChatGPT's Atlas, Perplexity's Comet, and Gemini in Chrome. "Front of house, it's making sure you're agentically ready. Back of house, it's how you can stack things [tools] in different ways to make you more efficient as a marketer," he said.
The research shows generative and predictive AI are already delivering measurable value. Agentic AI remains the frontier-powerful in concept but requiring organizational readiness around governance, data access, and tolerance for autonomous decision-making.
Your membership also unlocks:
