cryptographer - certified nerd god - founder @modulo_p_sa - ex-@snap - co-founder @strong_codes - has-been professor - trail runner - kayaker - aidjolat
Switzerland
Joined July 2010
- Tweets8.3K
- Following2.2K
- Followers3K
- Likes8.4K
Pascal Junod retweeted
Merci à l'équipe NoLimitSecu de m'avoir convié à un nouvel épisode de leur podcast!
nolimitsecu.fr/la-cryptograp…
Pascal Junod retweeted
#Podcast #Cybersécurité
Épisode #481 consacré au livre "Cybersécurité et hygiène numérique au quotidien", avec @GildasAvoine et @cryptopathe
nolimitsecu.fr/cybersecurite…
Pascal Junod retweeted
Jean-Pierre Serre (French mathematician, Fields Medalist) still making Wikipedia edits at 97 years old!
Pascal Junod retweeted
We are very excited to share our last research work: 𝐄𝐔𝐂𝐋𝐄𝐀𝐊, authored by Thomas Roche.
An electromagnetic Side-Channel Vulnerability in the ECDSA implementation of all Infineon security microcontrollers, notably impacting all YubiKey 5 Series.
ninjalab.io/eucleak/
Pascal Junod retweeted
The never-ending story. Rumor has it that both grooming detection and AI parts have been dropped, limiting the proposal to detection of known CSAM content. But even that is highly problematic as it does not work (easy to avoid), gives rise to a huge number of false positives...
Psst, dat voorstel om uw Whatsappberichten te controleren staat weer op de agenda. #ChatControl
datanews.knack.be/nieuws/eur…
Pascal Junod retweeted
Finally, a practical cryptographic side channel attack “in the wild”! Sort of. ninjalab.io/eucleak/
Pascal Junod retweeted
I really liked that this paper eprint.iacr.org/2024/1345 explains that the usage of PQC is more than TLS.. – at Valbonne, France
Pascal Junod retweeted
while Go adopts ChaCha8 at math/rand/v2, ARM's new RNG library uses MCG31 and MT19937
community.arm.com/arm-commun…
gitlab.arm.com/libraries/ope…
Pascal Junod retweeted
A long feature article on Grothendieck in the @guardian, with some actual reporting (which is quite rare on this subject) and photos I had never seen before. theguardian.com/science/arti…
h/t @OlivierReims and @coulmont
Pascal Junod retweeted
Today I’ve released a new simplifier for general MBAs: github.com/mazeworks-securit…
Pascal Junod retweeted
Anyone commenting on Durov's arrest should take note of three things: 1) Telegram has many different kinds of user communities across the world, for whom its particular mix of functionality, limited moderation and branding as a defender of free speech make it attractive.
Pascal Junod retweeted
Intel HW is too complex to be absolutely secure! After years of research we finally extracted Intel SGX Fuse Key0, AKA Root Provisioning Key. Together with FK1 or Root Sealing Key (also compromised), it represents Root of Trust for SGX. Here's the key from a genuine Intel CPU😀
Pascal Junod retweeted
Excellent blog post by @matthew_d_green explaining why Telegram is not really an encrypted messaging app. The TL;DR is that while Telegram offers the ability for two participants to engage in encrypted end-to-end communication, it is (1) off by default; (2) not easy to turn on (multiple menus / confirmations / restrictions); (3) requires the other participant to be online as well at the time of communication. In addition, metadata is exposed. None of this meant to be a knock on Telegram per-se, but calls out the fact that the people reporting on and discussing this issue are failing to use the correct nomenclature and not accurately representing what Telgram can do (particularly compared to how security professionals use that same terminology). blog.cryptographyengineering…
Pascal Junod retweeted
Et *par défaut* Telegram n'est pas une messagerie sécurisé,e et les groupes ne le sont juste pas peu importe les options...
Donc attaquer Telegram pour lutter contre les messageries E2E, c'est comme attaquer les trottinettes pour lutter contre les excès de vitesse...
Pascal Junod retweeted
It's amazing to me that after all this time, almost all media coverage of Telegram still refers to it as an "encrypted messenger."
Telegram has a lot of compelling features, but in terms of privacy and data collection, there is no worse choice. Here's how it actually works:
1/
Pascal Junod retweeted
Three decades after Shor's breakthrough paper, NIST has posted the first post-quantum cryptographic FIPS standards
203: ML-KEM nvlpubs.nist.gov/nistpubs/fi…
204: ML-DS nvlpubs.nist.gov/nistpubs/fi…
205: state-based hash nvlpubs.nist.gov/nistpubs/fi…
Pascal Junod retweeted
Trivia question: is it safe to use CBC-mode decryption as an *encryption* algorithm? (Assume an arbitrarily strong block cipher.)
Pascal Junod retweeted
A preprint of my paper "Deobfuscation of Semi-Linear Mixed Boolean-Arithmetic Expressions" (arxiv.org/abs/2406.10016) is now available.
This work extends algebraic MBA deobfuscation techniques to handle semi-linear MBAs - a class that existing techniques struggle with.
