Observed Signal · Sep 28, 2026 · Product Launch · Source: https://martechseries.com/feed/ · Impact: 2/5 · Sentiment: Positive
Omneky Launches TASTE BENCH for AI Creative Evaluation
Omneky, the autonomous AI growth platform, announced the launch of TASTE BENCH, an evaluation suite that tests how well image and video generation models convert creative briefs and brand assets into ready-to-run advertisements. The initial edition covers eight image models, 59 ad briefs, four brands, and five languages, generating 469 ads and 1,371 valid blind judge reviews. GPT Image 2.5 Sunburst achieved the highest ready-to-run rate at 54.2% (32 of 59), with average quality scores across models. A blind panel of AI judges from OpenAI, Anthropic, and Google scores outputs on eight quality dimensions, including brand fit, typography, and ad effectiveness. Eleven pass/fail checks cover exact copy, language, brand fidelity, safe-zone placement, and fabricated claims. Results serve as automated creative judgments, not campaign performance metrics.
The launch of TASTE BENCH provides a new benchmark for evaluating AI-generated ad creative, which is relevant to the AdTech and MarTech industry as it aims to systematize quality assessment across models. However, it is a specific tool launch from a single vendor, so its industry-wide impact is moderate.
Track Omneky Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Omneky launched TASTE BENCH, an evaluation suite for AI-generated ad creative.
- GPT Image 2.5 Sunburst achieved the highest ready-to-run rate at 54.2% (32 of 59).
- The benchmark covers eight image models, 59 ad briefs, four brands, and five languages.
- AI judges from OpenAI, Anthropic, and Google scored outputs on eight quality dimensions.
- The suite includes eleven pass/fail checks for brand fidelity and factual accuracy.
Connected Companies & Entities
1 Entity mapped“Omneky, the autonomous AI growth platform, introduced TASTE BENCH, an evaluation suite that tests how well image and video models turn creat...”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Optimizely Launches Marketing AI Models and Open Benchmark
Optimizely introduced a family of purpose-built, post-trained AI models for marketing tasks at its Opticon conference on September 1, 2026. The company claims these models deliver 10x cost efficiency compared to state-of-the-art large language models. Optimizely also launched Mark-Bench, an open-source benchmark evaluating AI performance across 285 tasks and 15 marketing functions. The Agent Platform scored a 67% all-pass rate on Mark-Bench default configurations versus 60% for Claude Code at half the cost. The models integrate with Mark-IQ, the platform's data layer, to automatically incorporate organizational context. The announcement follows the launch of Virtual Teammates the previous day, signaling Optimizely's push to scale agentic marketing beyond pilots. A global study of 2,000 B2B marketing leaders revealed 53% believe current AI tools struggle with emotional brand resonance.
AI Can't Spot AI Ads but Can Judge Human Creative Quality
Andrew Tindall, a senior leader at System1, explores whether AI can effectively evaluate human-made advertising, given that humans cannot reliably identify AI-generated ads. He cites four major studies over ten months, including NYU Stern's 'AI Advertising Paradox' (AI ads outperformed human ads by up to 19% without disclosure, but performance dropped 31.5% when disclosed) and a Taboola study with 500M+ impressions finding similar results for short-term metrics like CTR. In his own experiment, he ran ten well-known human-made ads through System1's AI testing tool, comparing its predictions to human-based ratings. Results show the AI predicted the Fluency Rating correctly 10 out of 10 times and the Star Rating 9 out of 10 times, with one refusal to score. AI performed best at extreme high and low performers, struggled with mid-table ads, and highlighted the importance of acknowledging uncertainty. Tindall concludes that AI is a probabilistic prediction machine, better used to supplement rather than replace human judgment in creative testing.
System1 launches AI creative screening tool
System1 introduced Test Your Ad Screen, an AI-powered creative screening tool trained on a dataset of 18 million human emotional reactions. The tool is designed to evaluate up to 100 long-form creative assets at once and provide predictions of commercial effectiveness within minutes, returning low/medium/high diagnostic bands rather than precise decimals. System1 says the model’s results match human tests in nine out of ten cases and that the tool will not return predictions when confidence is insufficient. Test Your Ad Screen is initially available via System1's online platform in the US and UK for long-form assets, with additional markets and media channels planned. The company positions the product as a pre-screen to its existing Test Your Ad consumer-testing solution.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.