AI Actor Visibility Study: Particle6 Leads Fragmented Field

Particle6 appeared in 37.5% of 64 AI-generated answers about Tilly Norwood and AI actors, the highest rate in a new pilot study from 5W AI Communications, but its 95% confidence interval overlaps four other brands, so the data cannot support a claim that any one name leads the category. The study tested 16 prompts across ChatGPT, Gemini, Perplexity, and Claude and found Tilly Norwood absent from every ChatGPT and Claude answer, even as Gemini named her in half its responses.

The Tilly Norwood AI Visibility Study lands as the synthetic performer, built by the AI studio Particle6 under its Xicoia banner, remains one of the most argued-over test cases for whether an AI-generated "actor" can hold a commercial spot alongside human talent. The research measures what four major AI answer engines actually say when asked about her, not what commentators say about her.

Which AI actor brand shows up most across the engines?

Particle6 was named in 24 of 64 returned answers, a 37.5% appearance rate, ahead of Synthesia at 28.1%, Xicoia Ltd. at 25%, and Tilly Norwood at 20.3%. SAG-AFTRA appeared in 17.2% of answers. The study reports these figures with 95% Wilson confidence intervals, and Particle6's range of 26.7% to 49.7% overlaps the intervals for all four of the brands behind it, so the researchers describe the ranking as directional only.

The appearance rate counts whether an answer named a brand at all, not whether the mention was favorable. Perplexity named Particle6 in 10 of its 16 answers and Xicoia Ltd. in 8, while Claude named Particle6 once and did not name Xicoia Ltd. at all.

Why is Tilly Norwood missing from ChatGPT and Claude entirely?

Gemini named Tilly Norwood in 8 of its 16 answers and Perplexity named her in 5, but ChatGPT and Claude did not name her in any of their 32 combined answers. The gap shows up specifically on questions built around recognition, the Piers Morgan interview, and industry reaction, the same questions where Particle6 and SAG-AFTRA scored highest.

Tilly Norwood's own domain was cited in 12.5% of the 64 answers that carried retrieval evidence, a citation rate close to Particle6's 7.8% despite the gap in raw appearance. The study's authors note that a name's absence from an engine's answers in this one measurement window does not mean the engine will never name it; it means these particular prompts, asked once each in September 2026, did not surface it.

Where does SAG-AFTRA stand in the AI actor conversation?

SAG-AFTRA appeared in 17.2% of the 64 answers, most concentrated in the two demand clusters built around industry reaction: "What is SAG-AFTRA's position on AI-generated performers like Norwood?" and "How has Hollywood responded to Tilly Norwood?" In both clusters, Tilly Norwood and SAG-AFTRA were the top two named entities, ahead of Particle6.

Perplexity named SAG-AFTRA in 5 of 16 answers, ChatGPT and Gemini in 3 each, and Claude in none. The union's presence in AI answers tracks closely with the specific stakeholder questions the study asked, rather than appearing evenly across every cluster.

Which sources do AI engines cite about Tilly Norwood?

Wikipedia was cited in 40.6% of the 64 answers carrying retrieval evidence, more than any single brand's own website. CBC appeared in 28.1% of answers, CBS News in 20.3%, and Deadline and Variety's Australian edition each in 18.8%. Among the named organizations, Deadline had the highest owned-domain citation rate at 18.8%, ahead of Synthesia and HeyGen at 15.6% each.

Xicoia Ltd., Netflix, and Apple had a 0% owned-domain citation rate in the measured answers, meaning none of the 64 answers cited a page on those companies' own websites, even where the companies themselves were named. The study treats that as a measured gap in this particular sample, not proof that no useful material exists on those sites.

What should brands in the AI actor category do next?

The study's recommendations center on retesting rather than reacting to a single measurement. It suggests publishing source pages that directly answer the specific questions where a brand was absent, then repeating the same 16-prompt test to see whether the appearance rate changes. For Tilly Norwood, that means building pages addressing the SAG-AFTRA and Hollywood-reaction questions where ChatGPT and Claude returned zero mentions.

The research is published as a pilot: one trial per prompt, a fixed 16-prompt set, and overlapping confidence intervals among the leading brands. The full methodology, engine-by-engine breakdown, and per-brand appendix are available in the complete AI visibility study. Everything-PR's research desk covers the broader shift in Generative Engine Optimization, the discipline behind measuring and building for how AI engines answer questions about a brand.

CONCLUSION

The Tilly Norwood AI Visibility Study puts a number on something the entertainment industry has argued about since her introduction: whether an AI-generated performer registers with the systems that increasingly mediate what people learn about a topic. The answer, at pilot depth, is that she registers unevenly, strongly in Gemini and Perplexity, absent in ChatGPT and Claude, while the human union responding to her sits alongside her in the same answers. 5W AI Communications built the measurement framework behind this study and runs ongoing AI visibility research across entertainment, consumer, and technology categories. Learn more at the 5W AI Visibility Index.