You might like
What if you could not only watch a generated video, but explore it too? 🌐 Genie 3 is our groundbreaking world model that creates interactive, playable environments from a single text prompt. From photorealistic landscapes to fantasy realms, the possibilities are endless. 🧵
🏉💡 What if your rugby strategy was powered by AI ? In my latest piece, I apply @karpathy's idea of a council of LLMs to generate battlecards to win the next game. 🧠⚡ Read here: sahbichaieb.com/rugby-llm-coun… #RugbyAnalysis #AI #SportsTech
sahbichaieb.com
Winning the Next Rugby Game with an LLM Council
Exploring how a council of AI can provide comprehensive tactical analysis for rugby teams, providing insights that boost performance and competitive advantage.
🏅 The Olympics may be over, but you can still score gold in AI security! 🔒 Check out the third article in my series on AI security. This one focuses on current tools to prevent prompt injection attacks in LLM systems: sahbichaieb.com/ai-security-to… #AI #Security #PromptInjection
sahbichaieb.com
AI Systems Security: Top Tools for Preventing Prompt Injection
Explore a selection of open-source tools that can help you scan your AI systems for vulnerabilities, mitigating the risks associated with prompt injections.
Notre 2ème #worshop "#IA et #data dans le #rugby international : la stratégie du XV de France" animé par @sahbichaieb 🤩 #BDAIP23 #IA #DATA @bigdataparis
Excited to share my new tool!🔒Analyze your profile & personality as seen by #ChatGPT. 🔥The goal is to raise awareness about #DataPrivacy . You can extract: -Personal Information -Life Summary -Personality Assessment -Hobbies/Interests github.com/sahbic/profile… #ResponsibleAI
github.com
GitHub - sahbic/profile-gpt: An AI-driven tool to analyze your profile and gain insights into how...
An AI-driven tool to analyze your profile and gain insights into how ChatGPT interprets your personality. - sahbic/profile-gpt
5 open-source frameworks to ensure privacy in your ML: 1. @openminedorg PySyft 2. @TensorFlow Privacy 3. @IBM adversarial robustness toolbox 4. @MetaAI CrypTen 5. @MSFTResearch SEAL
Very interesting point of view on the future of Cloud services by @bernhardsson
I wrote a blog post about how I think the cloud service stack will change in the next 5-10 years: erikbern.com/2021/11/30/sto…
AI and ML students across Africa, this one is for you 🌍. Do you want to take your research to the next level 🚀? Applications are now open for DeepMind Scholarships at Stellenbosch University (MSc and PhD) 🎓! mlai.sun.ac.za/dms @StellenboschUni @DeepMind
Can we use predictive models explainability for causal analysis ? I highly recommend this read, by @scottlundberg the author of SHAP approach bit.ly/34jJh2c
Happy to see the rise of open source tools for AI models security inspection. Meet Counterfit from @Microsoft #MLSec #AIRedTeam bit.ly/3ef9yoi
microsoft.com
AI security risk assessment using Counterfit | Microsoft Security Blog
Counterfit is a command-line tool for security professionals to red team AI systems and systematically scans for vulnerabilities as part of AI risk assessment.
Hi all. Did you know @GeorgiaTech has launched a new, first-of-its-kind School of Cybersecurity and Privacy (scp.cc.gatech.edu). We are seeking to hire the first School Chair to lead it. Check out the job ad: tinyurl.com/gt-scp-chair (I’m chairing the search; DMs are open)
Honored to start the conversation on Machine Learning security with @TinyCyber at the DevSecOps track @DevOpSummit bit.ly/3dGoHP1 #MLOps #AISecurity
Interesting thoughts on building a successful data-driven organization by @hassenchaieb
New blog post 📃 : Building blocks of a data-driven organization hassenchaieb.com/building-block… #DataScience #MLOps
Accuracy remains the first goal, adversarial robustness comes second for most applications.
Adversarial Training is Not Ready for Robot Learning deepai.org/publication/ad… by Mathias Lechner et al. including @orgrosu #DeepLearning #ComputerScience
deepai.org
Adversarial Training is Not Ready for Robot Learning
03/15/21 - Adversarial training is an effective method to train deep learning models that are resilient to norm-bounded perturbations, with t...
Interesting piece on the vulnerabilities of pickle files by @ESultanik. It shows how pickle files can be used to execute harmful code or corrupt your models. #mlsecurity #aisecurity #python blog.trailofbits.com/2021/03/15/nev…
blog.trailofbits.com
Never a dill moment: Exploiting machine learning pickle files
Many machine learning (ML) models are Python pickle files under the hood, and it makes sense. The use of pickling conserves memory, enables start-and-stop model training, and makes trained models...
Great news! This is going to be a great resource to learn how to secure ML workflows
We’re excited to announce the release of the second course in the #Private #AI #Series on March 16th at 5 PM GMT In this course, you’ll learn how to use foundational privacy technologies like homomorphic encryption and differential privacy. Learn more: blog.openmined.org/releasing-the-…
Really enjoyed this conversation and learned so much! There's good reason to believe neural networks look at very different features than we would have expected. Adversarial examples are still a bit of an unknown quantity and there is a lot of interesting research going on.
Adversarial examples have attracted significant attention in machine learning, but the reasons for their existence and pervasiveness remain unclear. Super deep dive with top researchers! Enjoy! @florian_tramer @wielandbr @ykilcher @ecsquendor youtube.com/watch?v=2PenK0…
United States Trends
- 1. Pond 245K posts
- 2. Jimmy Olsen 3,653 posts
- 3. Veterans Day 23.1K posts
- 4. Daboll 37.8K posts
- 5. Go Birds 13.7K posts
- 6. #GoPackGo 2,483 posts
- 7. McRib 1,888 posts
- 8. Downshift N/A
- 9. Akira 28K posts
- 10. #OTGala8 131K posts
- 11. #FlyEaglesFly 6,214 posts
- 12. Gorilla Grodd 1,978 posts
- 13. American Vandal 2,047 posts
- 14. Schoen 19.7K posts
- 15. Zendaya 10.6K posts
- 16. Mecole Hardman N/A
- 17. Biker 4,149 posts
- 18. Johnny Carson 1,093 posts
- 19. Harvard 34.1K posts
- 20. Nene Leakes N/A
Something went wrong.
Something went wrong.