Nethemba

Evaluating and mitigating the growing risk of LLM-discovered 0-days
Claude Opus 4.6, released today, continues a trajectory of meaningful improvements in AI models’ cybersecurity capabilities.

0-Days \ red.anthropic.com Nicholas Carlini*, Keane Lucas*, Evyatar Ben Asher*, Newton Cheng, Hasnain Lakhani, David Forsythe, and Kyla Guru *indicates equal contribution

Large-Scale Online Deanonymization with LLMs
We measure the capabilities of LLMs to deanonymize users online.

Large-Scale Online Deanonymization with LLMs We measure the capabilities of LLMs to deanonymize users online.

Venice Launches End-to-End Encrypted AI
We're introducing verifiably encrypted AI inference, offering both Trusted Ex*****on Environment (TEE) and End-to-End Encrypted (E2EE) models.

Venice Launches End-to-End Encrypted AI We're introducing verifiably encrypted AI inference, offering both Trusted Ex*****on Environment (TEE) and End-to-End Encrypted (E2EE) models.

Docker-based lab environment for the Attacking, Defending, and Leveraging AI class.

Integrates the BHIS AI-CTF (11 prompt injection challenges) with 9 hands-on Jupyter lab notebooks and an automated LLM red-teaming container.

GitHub - deruke/aisecops-labs: Collection of AI related labs to support AI Sec Ops podcast and related training classes. Collection of AI related labs to support AI Sec Ops podcast and related training classes. - deruke/aisecops-labs