Royce Moon

96 posts

Royce Moon banner
Royce Moon

Royce Moon

@compulsi0n

AI researcher, 0day enthusiast

nyc Katılım Mart 2022
29 Takip Edilen41 Takipçiler
Royce Moon retweetledi
Dino A. Dai Zovi
Dino A. Dai Zovi@dinodaizovi·
Hmmm... it seems like Mythos and Codex-5.5-Cyber haven't found *all* of the exploitable vulnerabilities just yet? It will be especially interesting to see the results in the browsers category this year, where Chrome and Firefox have been able to apply LLMs to fix what they can.
Xchg Labs@xchglabs

Pwn2Own hit max capacity for the first time in history. We unfortunately couldn’t get in. We submit our research to vendors. Look to our blog for technical breakdowns coming soon! #cybersecurity #0day #reverseengineering #exploitdev #infosec #xchglabs #pwn2own

English
2
6
30
4.6K
Royce Moon
Royce Moon@compulsi0n·
Our new paper introduces containment verification where we formally verify interactions between agents and the world via havoc oracle semantics, allowing for capability-agnostic, universal formal safety guarantees. We hope this can make a tangible path to trusted, safe AI!
Lav Varshney@lrvarshney

Royce Moon has done really incredible work on a containment layer for agentic AI. It ensures AI safety without having to dig into the internals or alignments of AI models, by formally verifying interactions with the world. Works whether considering superintelligence, super-duper intelligence, or anything beyond. arxiv.org/abs/2605.09045

English
0
0
2
180
Royce Moon
Royce Moon@compulsi0n·
Stay tuned for everything from @xchglabs 👀
International Cyber Digest@IntCyberDigest

‼️🚨 Pwn2Own Berlin 2026 just hit a wall. For the first time in 19-years, ZDI rejected dozens of working zero-day RCE submissions because organizers ran out of contest slots. Rejected hackers are now going public with PoC demos and direct vendor disclosures, breaking Pwn2Own's usual secrecy. ▪️ AI surfaces a massive wave of 0-day RCEs. ▪️ Submissions overwhelm ZDI past max capacity. ▪️ Slots run out. Researchers with working chains get rejected. ▪️ "Revenge disclosures" begin. ← we are here. Confirmed casualties so far: ▪️ @xchglabs : 86 vulnerabilities prepared (PyTorch, NVIDIA, Linux KVM, Oracle, Docker, Ollama, Chroma, LiteLLM, llama.cpp). All rejected. Now reporting directly to vendors with writeups dropping as patches land. ▪️ @ggwhyp : full-chain Firefox RCE on Windows. Rejected. Publicly demoed (HTML page → cmd.exe → calc.exe). Responsibly disclosed to Mozilla. ▪️ @yunsu_dev : working RCE chain, rejected. Submitting elsewhere. ▪️ @ryotkak : tried to register for 3+ weeks. ZDI confirmed "at maximum capacity, can't add extra contest days." Considered canceling flight and hotel. ▪️ @anzuukino2802 : Claude Code RCE PoC. Rejected. ▪️ @desckimh : 0-day RCEs in Ollama and LM Studio. Rejected. Reported impact: a community-estimated 150+ researchers tried to register. Accepted contestants are now being warned about collisions. Rejected vulnerabilities going to bug bounty programs may trigger pre-event patches that invalidate the work of those who got in. ZDI has not publicly addressed the capacity issue. The event still runs May 14-16 in Berlin.

English
0
1
2
419
Royce Moon retweetledi
Samuel Schapiro
Samuel Schapiro@samschapiro·
1/N Large language models (LLMs) have been widely adopted for closed-ended tasks like reasoning, but can they truly be creative? 📚 Excited to announce our new work — Combinatorial Creativity: A New Frontier in Generalization Abilities. 📝 Paper arxiv.org/abs/2509.21043
Samuel Schapiro tweet media
English
1
6
11
3.1K
Royce Moon
Royce Moon@compulsi0n·
@hthieblot building AI that can do AI research end-to-end autonomously and more
English
0
0
0
10
Hubert Thieblot
Hubert Thieblot@hthieblot·
@ Founders and builders with under 100 followers, what are you building? If it's cool, I'll invite you to our 40k sqft lab in SF
English
715
42
1.2K
163.4K
Royce Moon
Royce Moon@compulsi0n·
another banger
Spiral Works@spiralworks_ai

We’re excited to share that our recent paper, “Transformational Creativity in Science: A Graphical Theory,” co-authored by @samschapiro, @jonahablack, and @lrvarshney received the Best Short Paper Award at the 16th International Conference on Computational Creativity, organized by the @ACC_assoc. This work opens new avenues for developing AI systems capable of achieving breakthrough insights, mirroring the cognitive leaps seen in historical scientific revolutions such as Einstein’s theory of relativity, the heliocentric model, and the advent of non-Euclidean geometry. We’re grateful to ICCC for this recognition and look forward to continue developing a future where AI can contribute to robust, interpretable, and even transformative scientific discoveries. Paper link: arxiv.org/abs/2504.18687 #creativity #artificialintelligence #science #innovation

English
0
0
3
229
ruslan
ruslan@ruslanjabari·
every important company started with 99% of people who said it would never work… and 1-2 people who believed anyway find them and u survive dm me and you'll have one down
English
24
6
234
12.4K
Royce Moon retweetledi
Pedro Domingos
Pedro Domingos@pmddomingos·
Symbolic AI is overconstrained. Deep learning is underconstrained. When the two converge AI will be solved.
English
23
18
177
10.9K
aishik
aishik@xcellect·
The food here goes unreasonably hard
aishik tweet media
English
1
0
2
47
Royce Moon
Royce Moon@compulsi0n·
@AllanatrixQ hey my team and I are building a research OS for R&D. would love to chat!
English
0
0
0
65
Allan
Allan@AllanatrixQ·
Everything is open-source under Apache 2.0 papers written Models documented APIs defined All work grounded in functional utility—not theory alone
English
1
1
40
8.2K
Rohan Pandey
Rohan Pandey@khoomeik·
guy who thinks frontier models must be able to OCR sanskrit to qualify as AGI, yet is incapable of reading a single line of this manuscript
Rohan Pandey tweet media
English
72
65
1.5K
123.3K