Resources
Resources for trust, safety, and AI operations.
The Coolest Job At The Dinner Party: Why I Joined Cinder
AI as an Overwhelming Security Vector Requires a New Defensive Posture
Who gets to decide when AI is safe?
What Is AI Red Teaming?
AI Safety Is the Goal. Safeguards Are the Work.
The Real AI Safety Problem Is Already Here
The Best AI Starts with Human Judgment
AI Has Changed Our Tools, Not Our Standards
What Cinder Learned Red Teaming Alibaba’s Latest Open-Weight Model
Case studies
Customer stories
Real outcomes from the platforms trusting Cinder to protect their users.
Krea
Read case study
Cinder red teamed Krea 2 for enterprise-readiness before open-sourcing one of the most capable creative image models available today.
< 1%
NSFW rate across evaluated generations
Synthesia
Read case study
Synthesia extended its trust & safety stack with Cinder so enforcement keeps pace with how fast the product and the threat landscape are moving.
>97%
AI Moderation Accuracy
Black Forest Labs
Read case study
Cinder ran rigorous adversarial testing to help Black Forest Labs ship FLUX without compromising on safety.
>90%
Reduction in CSAM and NCII vulnerability
Character.AI
Read case study
Cinder’s agentic review reduced a significant percentage of cases, allowing moderators to focus on high-risk cases requiring human judgment.
>50%
Reduction of all queues