In specialist healthcare, finding out which engine describes your field accurately matters more than whether “AI mentioned us”.
Models evaluated: four frontier systems (GPT-4o, Gemini 2.0 Flash, Copilot and DeepSeek V3), tasked with open-ended clinical analysis of longitudinal dental cases.
For legal, medical and financial questions, a single conversation history is enough to give different users different advice. The strongest predictor is the topic.
Background: in high-stakes settings such as legal, medical and financial advice, even a single conversation history is enough to drive outcome differences between users, and prior work shows socio-demographic groups receive advice of differing quality.
39% of consumers think it is a good idea for their insurer to use AI to improve services, up from 20% — yet 36% are still less likely to buy from an insurer that publicly uses AI.
Survey scope: Insurity's 2026 AI in Insurance Report finds 84% of US consumers use AI tools at least occasionally and 27% use them daily.
When AI produces convincing reports and proposals at near-zero cost, signals built on costly effort stop working — the paper's substitute is a verifiable promise backed by damages.
Framework: generative AI is modelled as compressing the discernible headroom between what machines produce cheaply and what buyers can distinguish; below a critical headroom no separating equilibrium in production-side signals exists.
Across 32,155 clinical recommendations, models detected 71.1–89.6% but correctly referenced the source title only 3.6–29.7% of the time, and adherence ranged from 21.8% to 63.2%.
Scale: 3,418 clinical practice guideline documents from nine countries/regions and two international organisations over the past decade, spanning 24 specialties, from which 32,155 clinical recommendations were extracted with publication body, date, country, specialty, recommendation strength and evidence level.
The same facts can lead to different conclusions in China, California and Germany; models often answer correctly yet fail to cite the right jurisdiction.
Benchmark scale: three jurisdictions (China, California, Germany), 55 legal issues across contract, consumer, criminal, family and labour law, 6,149 instances grouped into 385 fact patterns, all reviewed by legal professionals.
Judging credit risk with a model alone ranks worse than a classic model — feed the classifier's key features and probability into the prompt and the metrics catch up.
Task: credit-card default prediction on the Default of Credit Card Clients dataset, comparing a fitted classifier, LLM prompting, and combinations of the two.
The same model's health information earns significantly different trust depending on whether it arrives as text, speech or an embodied interface.
Study 1 (N=21): across three health task types comparing sources from ChatGPT and Google, trust in health information from ChatGPT was significantly higher.
In heavily regulated sectors like UK iGaming, AI visibility is no longer decided by keyword density but by algorithmic trust — and compliance facts have to be machine-readable.
An empirical analysis of UK iGaming entities finds that compliance signals (such as UK Gambling Commission standards) act as authority multipliers for models like ChatGPT, Perplexity and Gemini once they are properly structured.