We already reported 7 vulnerabilities to ollama via huntr 🔥 The first one is CVE-2024-12886, report is now public even if the bug is not fixed yet... Please do not exposed your Ollama server, it is not the only bug we found... 🛡️ huntr.com/bounties/f115f…

Always a pleasure to welcomed at offensivecon 👌 This week will be intense! - 4 day browser fuzzing training - Conference & talks - Spend time with FuzzingLabs team - Drink beer with friends! - Maybe pwn2own 🔥 Thanks Binary Gecko ! 🙏

Boom! Mohand Acherir & Patrick Ventuzelo (@pat_ventuzelo) of FuzzingLabs (@fuzzinglabs) kick off Day Two in style by demonstrating their exploit of #NVIDIA Triton. They're off to the disclosure room to see if their exploit is unique. #Pwn2Own #P2OBerlin

We have another bug collision. Mohand Acherir & Patrick Ventuzelo (@pat_ventuzelo) of FuzzingLabs (@fuzzinglabs) exploited #NVIDIA Triton, but the exploit they used was known by the vendor (but unpatched). They still earn $15,000 and 1.5 Master of Pwn points. #Pwn2Own #P2OBerlin

Our partners and cybersecurity company Fuzzing Labs FuzzingLabs exploited NVIDIA Triton! Incredible work team. The only bad thing is that when they reported the bug NVIDIA already knew about its existence and was working on patching it. Congrats Patrick Ventuzelo, keep up the

🔥 First time at #Pwn2Own, first win — and $15,000 later, we’re just getting started. Last week at OffensiveConfessions, we trained, fuzzed, and pwned hard. Here’s the story 👇

In a video highlight from day two of #Pwn2Own Berlin, Mohand Acherir & Patrick Ventuzelo (@pat_ventuzelo) of FuzzingLabs (@fuzzinglabs) exploit the #NVIDIA Triton Inference server youtube.com/shorts/Xuol5l1…

🚀 Ready to go beyond static diagrams? This blogpost shares: ✅ Our workflow ✅ AI agents we built ✅ Real output examples ✅ Tips for using LLMs for proactive security 🔗 Read it here: fuzzinglabs.com/ai-threat-mode… 🧠 Built by FuzzingLabs — where automation meets offensive