Bar-El Tayouri
@turipo.bsky.social
@owasp.org can be really awesome sometimes
November 17, 2025 at 10:33 AM
@owasp.org can be really awesome sometimes
Everyone in AI security knows OWASP Top 10 for LLMs.
But most people don't realize there's a whole other framework they should be learning: MITRE ATLAS.
The difference? OWASP tells you what can break. ATLAS tells you how adversaries will break it.
🧵
But most people don't realize there's a whole other framework they should be learning: MITRE ATLAS.
The difference? OWASP tells you what can break. ATLAS tells you how adversaries will break it.
🧵
November 13, 2025 at 12:01 PM
Everyone in AI security knows OWASP Top 10 for LLMs.
But most people don't realize there's a whole other framework they should be learning: MITRE ATLAS.
The difference? OWASP tells you what can break. ATLAS tells you how adversaries will break it.
🧵
But most people don't realize there's a whole other framework they should be learning: MITRE ATLAS.
The difference? OWASP tells you what can break. ATLAS tells you how adversaries will break it.
🧵
Two years since Atom joined
@Mend_io
🎯
Security teams were drowning in alerts, fixing vulnerabilities that couldn't be exploited - like patching wiring never connected to power.
Atom's reachability analysis changed that: focus on code that actually runs. (1/2)
@Mend_io
🎯
Security teams were drowning in alerts, fixing vulnerabilities that couldn't be exploited - like patching wiring never connected to power.
Atom's reachability analysis changed that: focus on code that actually runs. (1/2)
November 11, 2025 at 2:17 PM
Two years since Atom joined
@Mend_io
🎯
Security teams were drowning in alerts, fixing vulnerabilities that couldn't be exploited - like patching wiring never connected to power.
Atom's reachability analysis changed that: focus on code that actually runs. (1/2)
@Mend_io
🎯
Security teams were drowning in alerts, fixing vulnerabilities that couldn't be exploited - like patching wiring never connected to power.
Atom's reachability analysis changed that: focus on code that actually runs. (1/2)
See how easy it was to create and edit some terminal simulation for my @blackhatevents.bsky.social talk. It's so silly but gif1337 is actually useful!
August 21, 2025 at 11:45 PM
See how easy it was to create and edit some terminal simulation for my @blackhatevents.bsky.social talk. It's so silly but gif1337 is actually useful!
if you are interested in AI security risks and you have a technical background this talk is for you
Bar-El Tayouri’s session, “Hacking and Securing AI Systems”, is now live!
Watch real-time demos of AI exploits and learn how to defend against them.
Don’t miss this #IntelliJIDEAConf session: youtu.be/Hj3RiAH0yCk
Watch real-time demos of AI exploits and learn how to defend against them.
Don’t miss this #IntelliJIDEAConf session: youtu.be/Hj3RiAH0yCk
August 21, 2025 at 10:13 PM
if you are interested in AI security risks and you have a technical background this talk is for you
Now on Security Boulevard 🎉
AI Red-Teaming - how and why.
AI Red-Teaming - how and why.
July 24, 2025 at 10:58 AM
Now on Security Boulevard 🎉
AI Red-Teaming - how and why.
AI Red-Teaming - how and why.
Time to switch AI security into proactive mode.
@blackhatevents.bsky.social AI Summit USA 2025 ☑️
@blackhatevents.bsky.social AI Summit USA 2025 ☑️
July 20, 2025 at 11:31 AM
Time to switch AI security into proactive mode.
@blackhatevents.bsky.social AI Summit USA 2025 ☑️
@blackhatevents.bsky.social AI Summit USA 2025 ☑️
Available in all sizes. Contact sales for live demo.
Comment if you want to kill a great idea in style 😎💀
Comment if you want to kill a great idea in style 😎💀
July 17, 2025 at 1:55 PM
Available in all sizes. Contact sales for live demo.
Comment if you want to kill a great idea in style 😎💀
Comment if you want to kill a great idea in style 😎💀
Sometimes social posts are great and sometimes it doesn't feel right.
I have a new Substack article, you are welcome.
turipo.substack.com/p/a-thought-...
I have a new Substack article, you are welcome.
turipo.substack.com/p/a-thought-...
A Thought on Physical vs. Cybersecurity
There’s a strange asymmetry between how we think about physical security and cybersecurity.
turipo.substack.com
June 26, 2025 at 3:01 PM
Sometimes social posts are great and sometimes it doesn't feel right.
I have a new Substack article, you are welcome.
turipo.substack.com/p/a-thought-...
I have a new Substack article, you are welcome.
turipo.substack.com/p/a-thought-...
MCP is overhyped.
It's a useful protocol, but the real story is about increasing agent autonomy. And the security risks that come with it.
Under-hyped protocols like x402 could be the real revolution.
Let's not lose sight of the bigger picture.
🔗 Article link in the first comment. 
It's a useful protocol, but the real story is about increasing agent autonomy. And the security risks that come with it.
Under-hyped protocols like x402 could be the real revolution.
Let's not lose sight of the bigger picture.
🔗 Article link in the first comment. 
June 12, 2025 at 9:20 AM
MCP is overhyped.
It's a useful protocol, but the real story is about increasing agent autonomy. And the security risks that come with it.
Under-hyped protocols like x402 could be the real revolution.
Let's not lose sight of the bigger picture.
🔗 Article link in the first comment. 
It's a useful protocol, but the real story is about increasing agent autonomy. And the security risks that come with it.
Under-hyped protocols like x402 could be the real revolution.
Let's not lose sight of the bigger picture.
🔗 Article link in the first comment. 
It was an absolute pleasure to speak about AI security, thank you @intellijidea.com for the invite ❤️
www.youtube.com/live/qg6tj8T...
www.youtube.com/live/qg6tj8T...
June 10, 2025 at 5:52 AM
It was an absolute pleasure to speak about AI security, thank you @intellijidea.com for the invite ❤️
www.youtube.com/live/qg6tj8T...
www.youtube.com/live/qg6tj8T...
Stop doomscrolling and check this epic AI security episode w/ @danielmiessler.bsky.social instead!
🎙️ I sat down with Bar-el Tayouri of Mend AI to talk about the future of AI security—and why it’s evolving faster than any category before it. We covered malicious models, agent discovery, and more. If you're building with AI, this one's essential.
youtu.be/538JTah7dTY
youtu.be/538JTah7dTY
A Conversation with Bar-El Tayouri from Mend.io
YouTube video by Unsupervised Learning
youtu.be
May 7, 2025 at 7:05 PM
Stop doomscrolling and check this epic AI security episode w/ @danielmiessler.bsky.social instead!