Advertisers use AI for optimization but hold back on creative work
TripleLift surveyed 200 advertising professionals in April 2026 and found a sharp divide in how the industry deploys AI. Seventy-three percent use AI for campaign optimization, but only 25 percent use it for creative production. The gap reveals where practitioners trust machines and where they insist on human judgment.
The report, published May 19, 2026, captures an industry caught between capability and caution. Most companies now run AI in some part of their operations, yet fully autonomous execution remains rare. The closer a task gets to the creative act, the more teams pull back.
Adoption is broad but shallow
Sixty-two percent of respondents said their company is still in the exploration or pilot phase of AI integration. That sits alongside a separate finding: 60 percent say they have a centralized AI strategy. The gap between having a strategy and trusting it is significant.
Fewer than 30 percent of respondents expressed high confidence in their AI strategy. Only 6.5 percent described themselves as "extremely confident," while 49.5 percent fell in the "somewhat confident" band. Teams are running pilots and early deployments, not institutionalized workflows with clear governance.
The four pillars of campaign execution
TripleLift frames campaign work around four distinct areas: media, measurement, audience, and creative. Each requires different AI systems. According to the report, the three most common AI combinations in use link media and measurement together, or all three without creative. Creative remains absent from nearly all integrated workflows.
The adoption numbers across those four pillars are stark. Fifty-nine percent use AI for audience targeting. Thirty-one percent report that budget allocation is handled entirely by AI. But creative production sits at 25 percent, and even within that group, over half deploy AI to scale and adjust existing assets rather than generate entirely new work. AI in creative functions as an assistant to human production, not a replacement.
The review tax
AI does not return time to practitioners as free hours. TripleLift found that 45 percent of respondents spend between one and four hours per week reviewing AI-generated outputs before they go live. Another 22 percent spend between four and eight hours on the same task. Six percent dedicate more than 12 hours weekly to ensuring campaign materials meet brand standards.
Twelve hours per week reviewing machine output represents a significant chunk of a working week. Even for the majority spending up to four hours weekly, the review process is a non-trivial commitment that compounds across teams and campaigns.
Trust is the bottleneck
When asked why they maintain human oversight in AI-driven workflows, 67 percent of respondents cited lack of trust in AI output as the primary reason. Technical errors and hallucinations came second at 61 percent. Brand safety checks registered at 41 percent, and data privacy concerns at 21 percent.
The trust figure is notable because it precedes specific, addressable problems. Brand safety and technical errors are the kind an improved model or better guardrails can reduce. A general lack of trust in output is harder to resolve. It reflects fundamental skepticism about whether AI systems can reliably handle contexts where a mistake - the wrong image, an off-brand phrase, a contextually inappropriate placement - can cause reputational damage before anyone notices.
Where AI is already delivering
The report is not uniformly cautious. In data-driven functions, practitioners are seeing measurable improvements. Early adopters report gains in speed (campaign launch time and optimization cycle duration), efficiency (eCPM and cost-per-click), and media metrics including click-through rate and conversion rate.
These are areas where AI's numerical and pattern-matching strengths map directly onto what advertising operations require. Bidding involves processing large auction data volumes faster than any human team can manage and adjusting in real time. Audience segmentation requires correlating first-party and contextual signals across large datasets to identify responsive groups. Both are tasks where algorithmic systems have clear advantages.
What remains contested is whether those backend efficiency gains actually translate into better outcomes for brands whose value depends on how they communicate creatively. The TripleLift data suggests the industry's working answer is that they do not - at least not enough to relinquish human creative control.
What creatives should know
For creative professionals, the numbers map directly onto job security and workflow changes. AI is automating the transactional and data-driven parts of campaign work. It is not yet automating the expressive parts.
The industry's stated priorities for future AI use rank inventory discovery and deal curation first, creative testing at scale third, and AI adaptations to existing creatives fourth. Even in the aspirational view, creative remains last. The industry envisions a future where AI adjusts and scales human-made creative rather than originating it.
That vision depends on creatives becoming faster at producing work that AI can then optimize and adapt. It also depends on creatives becoming more skilled at reviewing and correcting AI output - the "review tax" that currently consumes up to 12 hours per week for some practitioners.
For creatives considering AI training, focus on tools that augment rather than replace human production. AI Design Courses and Generative Art Training address the gap the TripleLift data identifies: the 25 percent adoption figure for creative AI reflects both limited tooling and limited confidence, not lack of demand.
Survey scope
TripleLift surveyed 200 respondents across six countries: 113 in the United States, 34 in the United Kingdom, 31 in France, 17 in Canada, 3 in Germany, and 2 in Australia. Respondents answered up to 30 questions each. The sample size is relatively small for a global survey, and the geographic skew toward US respondents is worth noting when reading cross-country inferences. The results are indicative of attitudes within TripleLift's professional network rather than a statistically representative cross-section of the entire advertising industry.
Your membership also unlocks:
