Observed Signal · Nov 9, 2026 · Technical Release · Source: Toms Marketing Ideas · Impact: 2/5 · Sentiment: Positive
AI Ad Generators Still Fail: How to Train Claude for Better Results
A creative director argues that current AI ad generators produce subpar results, but by training Claude with structured prompts and templates, one can generate effective ads. The article shares a method: research audience via Claude with a job-to-be-done framework, use templates derived from top-performing ads, and design ads in Figma with animation. The author claims Claude designed 24 ads for a client, Weglot, without guidance. The approach emphasizes simplicity: logo, headline, product shot. The article includes prompts for audience research and copywriting, positioning Claude as an 'ad factory' for marketers.
Article provides practical insights on using AI for ad creative generation, relevant to marketers exploring AI tools, but lacks industry-wide impact or breaking news.
Track Lovable Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Claude generated 24 ads for Weglot client in one chat session
- Article introduces a structured prompt for audience research using JTBD framework
- Author collected 100+ top-performing Google, Meta, and LinkedIn ads to create copywriting template
- Claude designs ads in Figma and animates them
- Validated ad format: logo, headline, product screenshot
Connected Companies & Entities
“Mentioned as an example of the same ad format.”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
AI’s Creativity Trade-Off for Marketers
The article argues that widespread adoption of generative AI in advertising is improving speed and scale but eroding creative originality. IAB data shows 83% of ad executives report using AI in the creative process (up from 60% in 2024) and 86% of buyers are using or plan to use generative AI for video creative. Academic and meta-analytic research (Wharton; a 28-study meta-analysis) finds AI can boost output quality and speed while narrowing idea diversity due to similar prompting. Real-world brand examples (Coca-Cola, McDonald’s) and IAB consumer research (January 2026) — where 82% of ad execs believed younger audiences liked AI ads but only 45% of consumers agreed — suggest overreliance on AI risks producing parity, weaker engagement, and brand backlash. The piece urges marketers to weigh short-term efficiencies against longer-term loss of distinctiveness.
AI Ads Match Human Performance in Groundbreaking Study
A large-scale study titled "AI Ads That Work: How AI Creative Stacks Up Against Humans" analyzed several hundred thousand live ads on Taboola’s Realize platform to measure the impact of generative AI on user behavior. The dataset included over 500 million impressions and about three million clicks. Using a sibling-ads methodology—pairing AI-generated and human-created ads from the same advertiser, campaign and day—the study controlled for targeting and timing to isolate creative effects. Raw click-through-rates were 0.76% for AI ads versus 0.65% for human ads, but after statistical controls overall performance was comparable. Perception mattered: ads visually identified as AI-generated performed worse, while AI ads that lacked typical AI cues achieved the highest engagement. The presence of clear human faces increased perceived authenticity; AI-generated creatives more often included such faces. The authors conclude AI is a tool that can match or improve creative outcomes when human perception and trust signals are considered.
AI Can Build Ads, But It Can't Run Them
The article argues that generative AI has democratized ad creative—making high-quality assets fast and widely accessible—but that the primary performance risks now lie in execution: compliance, targeting precision, delivery, and optimization. AI-produced creative can still fail in-market if it doesn't account for platform policies, regulatory nuance (notably in healthcare, finance, and politics), or channel-specific constraints. The author contends optimization systems require human judgment to balance measurable outcomes with brand intent and compliance. The piece predicts a hybrid model where AI accelerates production and pattern-finding, while humans lead governance, campaign structure, and cross-channel execution. Fyllo is identified as a data and advertising partner focused on regulated industries.
Receive all future signals for Lovable & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.
