91% of Marketers Use AI, But Only 50% Trust it to Act Autonomously
The report findings reveal where marketers are willing to let AI recommend, prepare, optimise and act, and where human oversight remains essential.
As artificial intelligence becomes increasingly embedded in advertising workflows, marketers are moving beyond the question of whether to use AI and towards a more complex question: how much decision-making authority should AI have?
Against this backdrop, StackAdapt has released The AI Delegation Gap, a global report examining how marketers are defining the boundaries between human judgment and AI-led decision-making across advertising workflows.
The report introduces a framework that maps marketers’ comfort levels across four stages of AI involvement: recommending actions, preparing actions for approval, acting within human-defined rules and making decisions autonomously.
Based on a survey of 500 marketing and advertising professionals across North America, EMEA and APAC, alongside insights from 187 StackAdapt customers, the research examines how trust, governance and operational readiness are shaping the next phase of AI adoption.
“The conversation around AI is shifting from how marketers use it to how much decision-making authority they’re willing to give it,” said Ryan Nelsen, Chief Marketing Officer at StackAdapt.
“That’s the AI Delegation Gap. As marketers build confidence in AI, they’re gaining a clearer understanding of where it creates value and where human judgment remains essential. How those responsibilities are divided will shape the next generation of advertising.”
Marketers are Drawing New Boundaries Around AI Decision-Making
The research shows a clear decline in confidence as AI moves from assisting marketers to making decisions independently.
- 90% of respondents are comfortable with AI recommending actions.
- 89% are comfortable with AI preparing actions for human approval.
- 78% support AI acting within human-defined rules and guardrails.
- Only 50% are comfortable with autonomous AI, even when its performance has been proven.
The findings point to a growing distinction between AI assistance and AI delegation. Marketers are increasingly willing to let AI analyse information, identify opportunities and prepare actions, but remain more cautious when those systems are given authority to act without human approval.
Organisations are Under Pressure to Operationalise AI Responsibly
While AI has become increasingly important to marketing teams, many organisations are still working to operationalise the technology effectively.
Nearly eight in 10 marketers (79%) feel pressure to increase their use of AI. At the same time, 59% believe leadership expectations around AI are ahead of their organisation’s current readiness.
Only 19% say their AI tools are fully integrated into marketing and advertising workflows, suggesting that adoption alone does not necessarily translate into mature AI operations.
The gap between executive expectations and operational readiness could widen as marketers move from experimenting with individual AI tools to integrating AI across campaign planning, execution, optimisation, and measurement.
Trust Depends on Transparency, Context, and Control
The report also identifies several factors preventing marketers from delegating greater decision-making authority to AI.
Brand risk is the leading concern, cited by 63% of respondents, followed by data quality concerns at 56% and lack of transparency at 36%.
The findings suggest that marketers are more likely to act on AI recommendations when systems can explain why a particular action has been suggested, connect recommendations directly to campaign objectives and provide mechanisms for testing, reversing or reviewing decisions.
This makes transparency and control increasingly important as AI moves from an analytical tool towards an active participant in campaign management.
AI is Becoming Part of the Decision Workflow
Reporting and performance analysis have emerged as the most common applications of AI among respondents.
- 77% use AI for reporting and summaries.
- 74% use AI for performance analysis.
However, the role of AI is expanding beyond simply producing reports. Marketers increasingly expect AI to identify potential risks, explain changes in campaign performance and recommend the next course of action.
This shift represents a move from descriptive automation to decision support, with AI increasingly helping marketers understand not only what happened but also what they should consider doing next.
AI Adoption Has Become the Industry Standard
The research also indicates that AI adoption has moved well beyond early experimentation.
More than nine in 10 marketers (91%) now use AI in marketing or advertising, while 88% say AI has improved marketing performance.
Adoption remains consistently high across regions:
- 91% in North America
- 90% in EMEA
- 92% in APAC
The consistency across regions suggests that the industry’s AI conversation is increasingly shifting away from adoption itself and towards questions of integration, governance and decision authority.
Human Oversight Remains Critical
Despite growing confidence in AI, marketers continue to reserve the strongest human oversight for decisions that can have long-term consequences for brands and businesses. These include budget allocation, targeting strategy, creative messaging and overall brand direction.
The report suggests that marketers are looking for AI systems that can operate within clearly defined boundaries, explain their reasoning and complement rather than replace human expertise.
“Organisations that succeed with AI won’t be the ones that automate the most,” added Nelsen. “They’ll be the ones that define clear decision boundaries, governance and accountability as AI takes on greater authority across advertising workflows.”
Six Principles for Responsible AI Delegation
To help organisations determine where AI can take on greater responsibility, the report identifies six foundational elements for responsible AI delegation:
- Context: AI needs sufficient business and campaign context to make relevant recommendations.
- Transparency: Marketers need visibility into how AI reaches its recommendations or decisions.
- Control: Teams should be able to establish clear boundaries around what AI can and cannot do.
- Testability: AI-led actions should be capable of being tested before being deployed at scale.
- Accountability: Organisations need clear ownership for AI-supported decisions and outcomes.
- Connected infrastructure: AI needs access to connected systems and data to operate effectively across the advertising workflow.
Together, these principles provide a framework for increasing AI’s role without removing the governance and human oversight that marketers continue to consider essential.
ALSO READ: Active Attention Involves Deliberate Decision-Making
