🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

With my co-organizers Shafi Goldwasser, Ankur Moitra and Dylan HadfieldMenell, I am organizing a workshop on LLM trustworthiness including security and safety at the Simons Institute on the Berkeley campus. Registration is free. Checkout the impressive lineup of speakers.

Thank you IBM Research for the Flock🪶 spotlight! Flock solves the problem of finding n non-colluding parties for a large class of multiparty cryptographic primitives! research.ibm.com/blog/flock-mul… 📰 Paper: usenix.org/conference/osd…

Don’t miss the Confidential Computing Summit! 🗓️ Register by 1/1 and save big 👉 hubs.la/Q030d0Zw0 🎤 Call for Papers open until 1/17 – share insights on #DataPrivacy & #SecureAI. 👉 hubs.la/Q030c_Z_0 #ConfidentialComputing #AI #trustworthyAI

📢 Call for Speakers Deadline Extended to 2/17 📢 You asked for more time, and we're delivering! Are you driving innovation in #ConfidentialComputing #DataPrivacy #GenerativeAI? Submit your proposal to speak here: hubs.la/Q032VnXH0 #SecureAI #CCSummit

Submit use cases or innovations related to confidential computing or confidential AI at the Call for Papers of the Confidential Computing Summit 2025 - due on Feb 17.

✨RL magic is in the air! Introducing DeepScaleR-1.5B-Preview—a fully open-source, 1.5B-parameter model trained with RL to surpass o1-preview for general math reasoning. 📜Blog: pretty-radio-b75.notion.site/DeepScaleR-Sur… 💻Github: github.com/agentica-proje…

Our students trained a 1.5B model that surpass o1-preview on math reasoning!

New and returning faces on The Cryptographers' Panel, a fan favorite #RSAC Keynote every year. Learn how the cryptographers see current events in cybersecurity and AI. spr.ly/60120yaPO April 29, 9:40 AM PT at RSAC 2025. Whitfield Diffie Ed Felten Raluca Ada Popa Vinod Vaikuntanathan

Announcing our evaluation framework for emerging cyber threats of AI @ Google DeepMind deepmind.google/discover/blog/…

Our team has open sourced our reasoning model that reaches o1 and o3-mini level on coding and math: DeepCoder-14B-Preview.

Thrilled to share our IEEE S&P '25 work "Myco 🌳🍄: Unlocking Polylogarithmic Accesses in Metadata-Private Messaging" with Deevashwer, Bhargav Annem | In Tokyo 🇯🇵, Raluca Ada Popa. We break a decade-old asymptotic barrier in cryptographic metadata-private messaging. eprint.iacr.org/2025/687👇

I am hiring a Software Engineer who has experience in program analysis and C/C++, as well as is generally a strong systems builder to join our AICodeSec team and our mission to use AI to fix vulnerabilities in code. If you are qualified, apply here: google.com/about/careers/…

I am hiring a cybersecurity expert at Google DeepMind to join our Frontier Security team in developing defenses that prevent AI misuse for cyberattacks. Apply if you are qualified, passionate about this, and can start in < 2 months. job-boards.greenhouse.io/deepmind/jobs/…

With #TalRabin, #WhitDiffie, #AdiShamir, Vinod Vaikuntanathan, Ed Felten at RSA conference, we discussed cryptocurrency, AI security and post-quantum cryptography.... in a coherent state with our past selves 🤓

The Myco talk is tomorrow at IEEE S&P. See you at 3:25 in Grand Ballroom A, Track 2, with the poster session right after.

Nicusor Dan (international math olympiad 1st place) just became president of Romania. 🇷🇴🥇 Terrence, no pressure

🚀 Introducing DeepSWE 🤖: our fully open-sourced, SOTA software engineering agent trained purely with RL on top of Qwen3-32B. DeepSWE achieves 59% on SWEBench-Verified with test-time scaling (and 42.2% Pass@1), topping the SWEBench leaderboard for open-weight models. 💪DeepSWE