Real-time detection of grooming, fake profiles and toxicity

Amanda is the proactive safety platform continuously monitoring your games and social channels to expose and stop unwanted behaviour before harm can happen

Detects grooming

in less than 20 messages

Accurate profiling

of age group and gender

Trained and tested

on typical chat lingo

Unbeatable performance

from layers of insight

Key Stroke Dynamics

Based on metadata from a chat (age and gender), Amanda is trained to determine gender and age group of the users. By analyzing how they type and their writing style Amanda can reveal if they really are who they claim to be.

Multi modal analysis

Amanda continuously analyze messages combining key stroke dynamics, stylometry, behavioral patterns, text and image.

Multi modal analysis

Amanda continuously analyze messages combining key stroke dynamics, stylometry, behavioral patterns, text and image.

Conversation risk score

The content and context of the conversations accumulate to a continuous risk score enabling real time prioritization of potentially dangerous conversations

Moderator superpowers

Up to 30x efficiency boost

Save time and money with targetet AI moderation.
Dramatically increase efficiency and empower the moderator to only focus on the reports that matter.

As text, imagery and sound are sendt back and forth, the content is continuosly evaluated by Amanda. Conversations with a risk score above the threshold are held for moderation and sorted by severity.

Ready to try?

pamanss

Let’s test your data, and see if there are conversations containing grooming, sexting, bullying and sharing of social handles your current moderation set-up has missed.