Do AI-generated ads perform as well as other ads?
Some AI-generated ads perform well in published tests, but comparisons produce different results depending on the sample and method.
In May 2026, Ipsos and Syracuse University tested 10 human-made ads and 10 AI counterparts with 3,000 US consumers. The originals scored 14% higher on the study’s short-term effectiveness index. The AI counterparts used reconstructed briefs, Gemini and Sora 2 without human creative intervention. That method differs from skilled people developing and refining work with AI. Ipsos’s study
The study is more than three months old. Its results concern the tools and work tested, not the limits of current production systems.
Does detecting AI make an ad less effective?
Detection alone does not establish how an advertisement will affect trust, memory or sales. Those are different outcomes.
In the Ipsos study, most viewers did not identify the AI ads confidently, yet the original ads performed better on average. The same research also found some effective AI ads. Recognition of the production method therefore did not explain the whole result. Ipsos’s findings
An artificial character, a product demonstration and a fabricated customer testimonial ask audiences to believe different things. Their implications depend on the claim and context, including whether the content could mislead.
What affects the quality of an AI-generated ad?
Quality depends on the idea, brief, source material, production choices and review, as well as the tools used.
AI can support creative exploration and adaptation, while people contribute knowledge of the audience and the brand. A production process can include repeated revisions, comparison of alternatives and checks against approved claims. The presence of those stages does not guarantee a strong advertisement, but it describes a materially different process from accepting an initial generation.
A polished image also answers a different question from whether the ad communicates the intended offer or builds useful associations with the brand.
How is an AI-generated ad’s effectiveness measured?
Effectiveness is measured against the advertisement’s intended outcome, using the same distinction between predicted response and observed behaviour that applies to other advertising.
Creative testing can examine recognition, response and likely impact. Campaign evidence can examine actions or sales, with suitable comparisons helping separate the ad’s effect from other changes. Neither a test score nor a lower production bill establishes profitable incremental demand on its own.
Evidence about a finished advertisement is more specific than a general verdict on AI. The conclusion remains bounded by the audience, market, channel and period measured.
Legal note: This answer provides general information, not legal advice. Seek advice from qualified legal counsel for your circumstances.