AI assistants confidently get SaaS product facts wrong

Name: AI assistants confidently get SaaS product facts wrong
Availability: InStock
Author: gissurthor

by gissurthor·Mar 30, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●●BangerZero to OneBig BrainSolve My Problem

Scans ChatGPT and Claude to catch lies about your SaaS pricing.

Strengths

•Identifies specific misclassifications like wrong category or missing integrations.
•Provides actionable content patches to fix AI knowledge gaps.
•Verifiable scoring loop tracks improvement over time.

Weaknesses

•$49/mo price point might be steep for early-stage startups.
•Relies on AI providers not changing underlying data ingestion methods.

Post Description

While building this, I found that ChatGPT, Claude, Gemini, and Perplexity regularly describe SaaS products with specific but incorrect details — wrong pricing tiers, deprecated features, misattributed integrations — with no uncertainty expressed. The tool runs a consistent set of buyer-style prompts across all four and shows verbatim output. Not trying to fix anything, just make it visible. One free scan. https://www.bersyn.com/