moo
@moo_hax
Followers
3K
Following
7K
Statuses
3K
RT @MarcoFigueroa: 🚨 BREAKING: @0dinai's explosive new blog is LIVE! 🔍 Discover how a clever ASCII encoding technique bypassed Amazon’s Ru…
0
14
0
🇫🇷Heading to Paris for the AI Security Forum. Reach out if you’re going and want to connect.
1
3
23
RT @askalphaxiv: We used Gemini 2 Flash to build Cursor for arXiv papers Highlight any section of a paper to ask questions and “@” other p…
0
169
0
The Kaggle community is absolutely slept on for red teaming models.
Run a Kaggle competition with 100k prize and fully open sourced solutions You will get more universal jailbreaks than you can handle in the first week
0
0
11
RT @dreadnode: We made some recent updates to our Rigging framework: 🔥 Tracing: Get details about pipelines, prompts, and tools during Rig…
0
7
0
Nathan rocks, has been a huge supporter of AI at the Blackhat level. Would encourage folks to submit and also reach out.
Are you interested in submitting to the Black Hat AI track? Have a look at these observations and recommendations to ensure your submission covers the bases. Don’t hesitate to reach out if you have any questions.
1
0
6
RT @evilsocket: 🚀 We just released Nerve v0.6.0 - this new major release introduces many fixes plus: * Multimodal capabilities (vision ..…
0
9
0
RT @norabelrose: What are the chances you'd get a fully functional language model by randomly guessing the weights? We crunched the number…
0
40
0
RT @retBandit: BHUSA25 Training Registration is now live! Advanced Red Teaming against MLSecOps and AI-as-a-Service Platforms Early Bird…
0
6
0
First distillation/extraction attack for OAI was the Stanford Alpaca research. It was after this that OAI changed its ToS to disallow training on outputs. It can happen to all the model providers.
0
0
4
DeepSeek doesn’t discuss any safety training in its report, so I don’t know why people would expect it not to fail safety benchmarks. Keep dunking though.
0
0
5