I made "Awesome LLM Security" repo for providing a curation of awesome tools, documents and projects about LLM Security. Let's make LLM secure together! github.com/corca-ai/aweso…

LLMs red-teaming LLMs is an old idea from over a year ago. LLMs red-teaming LLMs with XSS is a new one! hackstery.com/2023/07/10/llm…

My new bi-weekly newsletter on AI Security and Safety - Real Threats of Artificial Intelligence :) hackstery.com/2023/07/20/rea… #ai #aisecurity #llm #hacking #it

AssemblyAI how can I submit some details on the Prompt Injection vulnerability in LeMUR? I only see "contact sales" option on your website #llm #ai

If you ever wanted to poison a training data set, here's your chance. pcmag.com/news/elon-musk…

I've published #2 release of my AI sec. & safety newsletter! Indirect Prompt Injection magic, Prompt Injection -> RCE, securing AI pipelines and many more here! hackstery.com/2023/08/01/rea… Please share & subscribe 😃

Blog on indirect prompt injection using YouTube video in LeMUR by AssemblyAI. No real impact on security, but on the other hand interesting demo on how you can inject prompts with third party resources. hackstery.com/2023/08/11/ind… - like & share 🙂

Here it comes - #3 release of Real Threats of Artificial Intelligence! If you're into LLMs, AI hacking and stuff like that you should subscribe ;) hackstery.com/2023/08/15/rea… #ai #llm #aisecurity #aihacking #cybersecurity #hacking

Next issue of the Real Threats of Artificial Intelligence newsletter! Inside you will find two excellent articles on software supply chain security in Machine Learning and many more: hackstery.com/2023/08/31/rea… #ai #LLM #CyberSecurity #MachineLearning

Real Threats of Artificial Intelligence newsletter is here! hackstery.com/2023/09/19/rea… Enjoy your read and subscribe to the mail list! 🙂

I've compiled a payloads list for attacking LLMs - check it out. Those payloads are useful for testing for some of the vulns from OWASP Top10 for LLM. github.com/mik0w/pallms #LLM #AISecurity #LLMSecurity #Cybersecurity #Hacking

Patch your PyTorch deployments: oligo.security/shelltorch "ShellTorch allows access to proprietary AI models, insertion of malicious models, leakage of sensitive data – and can be used to alter the model’s results or to execute a full server takeover." #CyberSecurity #AI

Check out this LLM wargame: tensortrust.ai! Super cool, kudos @tensortrust,Sam Toyer and the rest of the team. #AI #LLM #Hacking

the 6th release of Real Threats of Artificial Intelligence is here! check it out: hackstery.com/2023/10/05/rea… if you have some interesting resources that you think would fit into this newsletter, drop me a dm or mail. #LLM #AISecurity #Hacking

Thank you for sharing LLM Security!

Short post on how I was able to get access to OpenAI API keys, GH keys and whole bunch of hardcoded passwords through misconfigured MLOps/LLMOps (😅) software: hackstery.com/2023/10/13/no-… PS why is there Ops for everything these days? #mlsecurity #aisecurity #llm #ai #hacking

Tip: When you really like a tweet, click on the author's profile from the expanded post It's worth 24 likes🤯 That's almost as much as a reply, and you can also do it multiple times Thanks for coming to my TED talk

when you're trying to find some niche information on the Internet, but Medium "bloggers" have already managed to pollute Google's search results with their useless GPT-generated articles