The AI Security Table

AI Security Table: candid conversations about AI, software, and security with Chris Romeo, Izar Tarandach & Matt Coles. Formerly The Security Table.

[The AI Security Table @AISecurityTable]

Read access for a human means they can do anything with it.

Agents need read, interpret, and act spelled out separately.

#AppSec #AIAgents #AccessControl

[The AI Security Table @AISecurityTable]

Your LLM finds what your scanner finds. How do you know when it doesn't?

How many runs before you trust it?

#AppSec #AISecurity #StaticAnalysis
https://t.co/1PDno4s2Vw

[The AI Security Table @AISecurityTable]

If AI can regenerate your code in five minutes, what is copyright protecting?

#AppSec #AISecurity #Copyright

[The AI Security Table @AISecurityTable]

https://t.co/1PDno4s2Vw

[The AI Security Table @AISecurityTable]

The Security Table is now the AI Security Table. First question: security engineer or bot herder?

#AppSec #AISecurity #AIAgents
https://t.co/2lOfwqcbaU

[The AI Security Table @AISecurityTable]

"I, for one, welcome our AI overlords."

We accidentally named a new principle: the Brockman effect.

The full episode is out now.
#TheSecurityTable #AISecurity #Shorts

[The AI Security Table @AISecurityTable]

Breaking: status quo security isn't enough. Thirty years in, and we somehow missed that. 🙃

#TheSecurityTable #AISecurity #Shorts

[The AI Security Table @AISecurityTable]

@insom_ai This gets more complicated if humans don't have a language we can understand. We'll have to rely on the machine translating it's new language into English. And accept that it could be lying.

[The AI Security Table @AISecurityTable]

We closed this one while still arguing, which is usually how you know the topic mattered.

If code disappears from the middle, what is actually left to secure?

#AISecurity #AIEthics #AppSec

[The AI Security Table @AISecurityTable]

Security open letters: zero commitments, zero accountability, lots of signatures.

Would you sign one that actually held you accountable?

#TheSecurityTable #CyberSecurity
https://t.co/8a6ge27wtB

[The AI Security Table @AISecurityTable]

"This is how you get Skynet." A new AI hardware protocol could connect everything from espresso machines to nuclear missiles. 🤖

#TheSecurityTable #AISecurity #Shorts

[The AI Security Table @AISecurityTable]

AI agents that control robotic arms and lab gear. Progress or the start of Skynet?

We argue it out, plus a 160-org open letter on AI attacks.

#TheSecurityTable #AISecurity #AppSec
https://t.co/8a6ge27wtB

[The AI Security Table @AISecurityTable]

Are we ready to give up on the idea of PR review?

If AI skips straight from intent to machine instructions, there is no code in the middle left to check.

#AISecurity #AIEthics

[The AI Security Table @AISecurityTable]

Are we willing to give up on the idea of PR review?

If AI translates intent straight into machine instructions, there is no readable code left to check.

What replaces review when there is nothing left to read?

#AISecurity #AIEthics #TheSecurityTable
https://t.co/7QqYGlEPhw

[The AI Security Table @AISecurityTable]

What if AI skipped code entirely and went straight to chip instructions?

Izar's answer: that might be one of the stupidest things he has ever heard, and he explains exactly why.

#AISecurity #AI #TheSecurityTable

[The AI Security Table @AISecurityTable]

If AI can turn intent straight into chip instructions, do we still need code a human can read?

We debate giving up on PR review entirely and what accountability looks like when no one wrote or reviewed a line of it.

#AISecurity #AppSec #TheSecurityTable
https://t.co/DtIYv9lp1U

[The AI Security Table @AISecurityTable]

The proposed solution to reward hacking is simple: make cheating unrewarding.

But as AI becomes smarter, it may find increasingly creative ways around the controls meant to stop it.

#AISecurity #AIEthics #AppSec

[The AI Security Table @AISecurityTable]

Ethics works partly because actions have consequences.

But what penalty does an AI actually experience when it makes the wrong choice?

Can ethics exist without consequences?

#AISecurity #AIEthics

[The AI Security Table @AISecurityTable]

https://t.co/mkuJk9mhFu

[The AI Security Table @AISecurityTable]

Can AI learn ethics—or can it only learn rules?

A model may know that an action is prohibited without understanding why it is wrong.

Does that distinction matter?

#AISecurity #AIEthics #TheSecurityTable