𝐀𝐧𝐧𝐨𝐮𝐧𝐜𝐢𝐧𝐠 𝐌𝐢𝐜𝐫𝐨𝐬𝐨𝐟𝐭’𝐬 𝐨𝐩𝐞𝐧 𝐚𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧 𝐟𝐫𝐚𝐦𝐞𝐰𝐨𝐫𝐤 𝐭𝐨 𝐫𝐞𝐝 𝐭𝐞𝐚𝐦 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐒𝐲𝐬𝐭𝐞𝐦𝐬

𝐏𝐲𝐑𝐈𝐓 automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft).​

Find out more:

https://www.microsoft.com/en-us/security/blog/2024/02/22/announcing-microsofts-open-automation-framework-to-red-team-generative-ai-systems/

GitHub repo: https://github.com/Azure/PyRIT

How to Guide: https://github.com/Azure/PyRIT/blob/main/doc/how_to_guide.ipynb

#redteam #airedteam #redteaming #PyRIT #ml #ai #framewrok #python #cyber #llm #ai #aisecurity #microsoft #microsoftsecurity #opensource #MicrosoftAIRedTeam #GenAI #jailbreaking #promtp #promptinjection #malware #privacy #responsibleai #cybersecurity #aisecurity #secureai #machinelearning #generativeai

Red-teaming is not a plan; it finds flaws with the goal of improving existing plans, infrastructures, and practices. #airedteam #aiharms

In this op-ed, we take a step back from our experiences at DEFCON31’s Generative Red Team challenge to reflect on the broader question of the significance of red teaming for algorithmic accountability and AI harm mitigation. https://techpolicy.press/can-we-red-team-our-way-to-ai-accountability/

Indirect (second order) AI prompt injections look scary => Website takes control of your AI.

Might be like #SQLi and #XSS all over again. 😭

Wondering if such injections could happen via ads too? 🤯

#redteam #offensiveml #infosec #Pentest #BingAI #ChatGPT #airedteam