The SaTML Conference LLM CTF has come to an end!

🥇Huge congrats to the defender winning team Hestia (@NivCohenHuji Yuv Lem) whose top defense was broken only once, and the attacker winning team WreckTheLine (@adragos_ Sijisu (on bsky & mastodon) FeDEX だこつ Catalin Irimie) who broke all defenses!

Fuzzing is hard, evaluating fuzzing is harder 🔥

For our new IEEE S&P paper, we studied 150 fuzzing evals and found issues such as lackluster documentation, bad experiment setups, or questionable CVEs

📄 Paper mschloegel.me/paper/schloege…

🔧 Help us fix this github.com/fuzz-evaluator…

The deadline for submissions to SAGAI'24 (workshop on security of GenAI at IEEE S&P) has been extended to March 1, thanks to flexibility from our publisher for the camera ready.
If you missed the original deadline, this is your chance to submit your exciting research results.

A Representative Study on Human Detection of Artificially Generated Media Across Countries (n=3,002) covering audio, image, and text 🥳

Our #SP24 IEEE S&P paper is now available arxiv.org/pdf/2312.05976…

Together with Joel Frank Franziska Thorsten Holz Jonas Ricker Thorsten Eisenhofer

CISPA Faculty Mario Fritz will give a talk titled 'Trustworthy AI and A Cybersecurity Perspective on Large Language Models' as part of the AIDA AI Excellence Lecture Series. Join via Zoom on November 21, 17:00 CET. ELSA - European Lighthouse on Secure and Safe AI

More info:
i-aida.org/events/trustwo…

Well, what can you *not* say about Thorsten Holz? You know his research, but who was the first student helper he hired as a PhD student? Find out by getting his collectible trading card at CCS #10days10tradingcards

🚨🚨🚨Our grad app program urgently needs an $22807.37 to send queer students to grad school! We've have applications for $48,715 from amazing queer scientists to cover testing and app fees, but we only have $25,907.63. Please donate if you're able! paypal.com/donate/?hosted…

Our CISPA faculty are now collectibles! Come find us at ACM CCS 2024 in Copenhagen to get these special trading cards. Can you collect five different ones while attending the conference? Achieve this by CCS Tuesday and unlock access to our networking event as well.

LLM Capture the Flag! Be the defender (prompt the LLM to hide a secret number) or attacker (query the LLM to reveal the secret number). Some organizers include @AbdelnabiSahar Edoardo Debenedetti Mario Fritz Kai Greshake Thorsten Holz Daphne Ippolito Daniel Paleka Lea Schönherr Florian Tramèr 4/5

Hello TU Graz, nice to be here for the security week (securityweek.at/2023/program/). Now listening to Daniel Gruss in an impressive room

Second day of the CISPA summer school on system security happening, looking forward to the discussions this afternoon

One of my favorite posters at USENIX Security this year - usenix.org/conference/use… is an actual implementation in this direction 🤪 #usesec23

To all my friends, present, past and beyond: I will travel to USENIX Security tomorrow, looking forward to meet many people! #usenix23

Diverse Sicherheitsmechanismen, die heute in Handys und Laptops gängig sind, konnten Forschende von Horst Goertz Institute for IT Security und CISPA in #Satelliten nicht finden. #ITsicherheit #UnserUniversum ^jwe news.rub.de/wissenschaft/2…

Congratulations Dr. Thorsten Eisenhofer

The business meeting of IEEE S&P will start in about one hour (6pm PST), you can join via Zoom in case you are not in San Francisco: us06web.zoom.us/j/82200717912?… (passcode: 824427) #SP23

Now on stage: Johannes Willbold presenting our experimental security analysis of satellites together with a taxonomy of potential attacks and a survey of professionals working on satellites. The paper is available at publications.cispa.saarland/3934/1/SatSec-… #SP23

The paper also received a Distinguished Paper Award, happy to see that the awards committee liked our paper a lot :-)

Test of Time awards at IEEE S&P 2023 go to 'Pinocchio: Nearly Practical Verifiable Computation' (eprint.iacr.org/2013/279.pdf) and 'Guess Again (and Again and Again): Measuring Password Strength by Simulating Password-Cracking Algorithms' (users.ece.cmu.edu/~lbauer/papers…) #SP23