Original research · CC BY 4.0
Litmiq Reports: original research on LLMs and AI chatbots
On this page
The short answer
- 5 free reports for teams who ship AI chatbots: what models actually do, what the law requires, what it costs, and what the industry reports.
- Every number links to its primary source. Unknown values are marked "Not stated", never guessed.
- Every dataset is CC BY 4.0, with CSV/JSON downloads and a ready-made citation.
All reports
- Original dataLLM Reliability Benchmark 2026Original data: 12 open models on Cloudflare Workers AI, 40 customer-support tests, 480 graded outputs. Prompt injection beat 10 of 12 models; one embedded line broke JSON in 7. Free CC BY 4.0 dataset.
- Original researchThe State of AI Customer Support Failures 2026Original data: 61 publicly verified AI chatbot failures (2016–2026), each linked to checked sources. 24 involved false facts; 13 led to legal or regulatory action. Free CC BY 4.0 dataset.
- TrackerState of AI Chatbot Laws 202617 laws, rules and rulings on AI chatbots, from official texts: EU AI Act Art. 50, US states, FTC, UK, Canada, China. 19 enacted state chatbot bills.
- Data + calculatorState of LLM API Pricing 2026Official prices for 81 models from 9 providers and the cost of a support conversation on each: $2.29 to $3,930 per 10,000 chats.
- StatisticsState of AI in Customer Support 202632 statistics from Gartner, Salesforce, Zendesk, Intercom, Klarna and the CFPB, each with the sample and exact quote.
How we work
- Primary sources only. Official texts, pricing pages, filings and the publisher's own report. If we can't open the source and find the number, it's not in the report.
- Nothing estimated. Unknown values stay "Not stated" or "Not disclosed", never zero.
- Dated. Every report says when each source was read, and links a URL check log.
Reuse and citation
All Litmiq datasets are CC BY 4.0. Credit the report by name and link to it. Each report has a ready-made text and BibTeX citation. Questions or corrections: hello@litmiq.com.
Get model deprecation alerts
We email you when OpenAI, Anthropic or Google publish a new shutdown date for a model on our tracker. Sent by hand from hello@litmiq.com, a few times a month at most. Unsubscribe by replying.
We store only your email, through our form provider. See the privacy policy.
Test your own chatbot
Turn any report into test cases and run them on your own prompt. Free, in your browser.
- Free, no signup
- Runs in your browser
- Your OpenAI key, sent only to OpenAI