Observed arrival · 2026-10-09
Supa Brain checks what customer-service chatbots get wrong
Supa Brain says it tests a website’s AI chatbot with customer questions and checks the answers against the site’s own pages.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Businesses using website customer-service chatbots
- Worth noticing
- The page names seven failure patterns, including wrong pricing rows, answers that contradict site pages, and chatbots that loop or go silent.
Field notes
The page frames the work as checking answers against a business’s own published pages, rather than simply judging whether a chatbot sounds plausible. Its examples span both factual mistakes—such as selecting the wrong rate—and friction that can block a customer, including repeated consent screens and requests for personal details before answering a public FAQ. Intercom Fin, Zendesk AI agents, and Tidio Lyro appear in the platform list.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
◎NicheUnusually specific use
One card from the complete issue