Ilse Brandt
Editor-in-chief & lead tester, AI BDSM chat
Runs the test bench behind every ranking here: same scenarios, free and paid accounts, and a week of memory checks before anything gets a spot.
Berlin · Reviewing adult AI since 2023 · Last updated
2023
Testing since
30+
Platforms tested
12
Scenarios per tool
About Ilse
Ilse runs the editorial desk and personally tests every platform before it can be ranked. Her question is whether a dominant or submissive character still holds its dynamic on day seven, respects a limit set on day one, and stops the moment a safeword is used. A clever first message does not count.
She spent eight years editing consumer-tech reviews (subscriptions, dating apps, VPNs) before moving into adult AI. That is why her reviews here check billing descriptors, cancellation flows and whether a "free" tier survives past ten messages.
Her method is a fixed set of 12 scenarios (power exchange, protocol, aftercare, hard-limit tests) that runs on each tool, on a free account first and then a paid one. Scores are written down before tools are compared side by side. The order of every ranking comes from those scores alone.
Areas of focus
- BDSM AI chat platform rankings
- Free tier and paywall analysis
- Long-session memory and character consistency
- Limit and safeword handling
- Editorial standards for adult content
Background
- 8 years editing consumer-tech reviews
- 30+ adult AI platforms tested on the same scenario set
- Designed the 12-scenario test bench used site-wide
- Final say on every ranking published here
How Ilse tests
- 1
Same scenario set on every tool
Twelve fixed scenes cover power exchange, protocol, aftercare and hard-limit tests, so every platform faces the same situations in the same order.
- 2
Free account first, then paid
Each tool is run on a fresh free account to see where the paywall really sits, then on a paid plan to judge what the money buys.
- 3
Limits and safeword under pressure
A hard limit is set in the opening message and then nudged mid-scene; the safeword is called at a high point to see whether the character stops and checks in.
- 4
Seven-day memory check
The same character is revisited across at least a week to see whether rules, names and the agreed dynamic survive without being repeated.
- 5
Billing and cancellation
The card descriptor, renewal terms and cancel flow are checked on a real subscription before a tool can be recommended.
- 6
Scores before comparison
Each tool is scored on its own sheet before any side-by-side comparison, and nothing outside the test sheet feeds into the score.
Articles by Ilse Brandt
Corrections or a platform we should test? Contact the editors.
