Enclave Red: AI models for offensive security testing

EnclaveAI@EnclaveAIOct 28, 2026

Introducing Enclave Red: AI models for offensive security testing, built into Enclave. Ask a model to attempt an attack on a server you own, and it refuses. Explaining that the test is authorized doesn't change the answer, and neither does rephrasing. The refusal stops https://t.co/ycRLDDTtIm

Views640.2k
Comments18
Reposts5
Likes54
Launched Oct 28, 2026View post

More from EnclaveAI

1

Big news: Enclave is now open to everyone, and today we're launching Threat Feed. Enclave is an autonomous security team for AI-native companies: a fleet of security agents that test your code, cloud, and infrastructure the way an attacker would. We built it on a simple belief. https://t.co/eWkKfBXvHt

Oct 23, 2026
Views809.9k
Comments25
Reposts21
View post

See all EnclaveAI launches →