Frequently asked questions.
Short answers first. Follow the links for detail.
The basics
What is Artixol?
Artixol is a research tool that builds a living audience of AI personas and runs surveys, message tests, shocks, interviews and rumor studies on them. Personas remember what they answered, so you can study the same people again.
What is a synthetic audience?
A synthetic audience is a group of simulated people, each with a role, background, personality and attitudes, generated by AI from a brief. In Artixol you describe who you want to study and it staffs 1 to 200 personas.
How is Artixol different from a chatbot?
A chatbot answers one question in one voice and forgets. Artixol staffs many different personas, collects structured and comparable answers, and writes the result back into each persona's memory so the next study builds on it.
What can I study with it?
You can run a Survey, a Message test, a Shock (a world event), an Ask (open question or 1:1 interview), a Rumor (word of mouth along relationships) and a Lab A/B experiment on cloned samples.
How long does a study take?
Most studies on an audience of a few dozen personas finish in minutes. Larger audiences take longer because each persona answers individually.
Accuracy and trust
How accurate are the results?
Artixol results are experimental model output, not a census or a forecast. They are best used to find where to look, which messages to test with real people and which segments to recruit. See the methodology page for limits.
Can I replace my survey panel with Artixol?
No. Use Artixol to sharpen the questionnaire, narrow options and decide where to spend a field study. Keep human research for decisions that need real evidence.
Do personas just agree with everything?
Personas are prompted to answer from their own job, daily life, languages and memories, and replies must vary by person. We also record structured choices, which makes disagreement visible in the splits.
Where do the personas come from?
They are generated from your brief by a language model. If you name a country or city, names and languages follow that culture. If you do not, the default is the United States, including coherent migrant communities.
Using the product
How many personas can I create?
You can create between 1 and 200 personas per audience today. Larger samples are on the roadmap.
What does 'memory' mean?
Memory is what a persona remembers from prior studies: episodic memories and topic attitudes (a 0 to 10 score with a note). Memory is what lets a follow-up study see earlier answers.
What is the difference between 'same people + memory' and 'same people, fresh minds'?
'Same people + memory' clones an audience and keeps what each persona believes. 'Same people, fresh minds' keeps the profiles and clears study memory and attitudes, which is what Lab A/B uses for fair comparison.
Are places and relationships required?
No. Places (city and neighborhood grounding) and relationships (family, peer and colleague ties) are opt-in and off by default. Rumor requires relationships.
Can I export my data?
Yes. Study results export to CSV, JSON and PDF.
Data and security
What happens to the text I paste in?
Messages, questions and briefs are sent to the configured language model provider to generate persona responses. Details are on the security page and in the privacy policy.
Do you train models on my data?
Artixol does not train models on your studies. Your content is sent to the configured language model provider to generate responses, and that provider's data terms apply. Review the security and privacy pages.
Can I delete an audience?
Yes. Deleting a project removes the audience and its studies.