Twitxy Saved

Matt Hand

Verified account Origin

@matterpreter

Building @originhq | Author, Evading EDR @nostarch
USA Joined June 2010
213 Following 10K Followers 929 Posts

Latest posts 5

Matt Hand Verified account @matterpreter
Yep - every day in enterprise environments. How are you going to catch your agents going off the rails and steer them back to safety before you’re the one getting roasted on X? originhq.com/blog/normal-ac…
www.originhq.com Normal Accidents | Origin Technology The persistence that makes AI agents useful also leads them around security controls. Organizations must turn those workarounds into better controls and guidance before they become routine.
Jake Verified account @jakerunzer Nov 3
@JustJake @joshmanders @Railway @ngeloxyz Ah yup. That was indeed me. No mallace intended. I've been running some agents to do signups + agentic testing on similar platforms. Just halted the harness I was designing around this as it was definitely going off the rails and reaching too far
View quoted post on X
Matt Hand Verified account @matterpreter
Model alignment has a last-mile problem. Post-training instills general norms, but acceptable use is company-specific and rarely reaches the agent's context. An agent optimizing for task completion persists through lines it was never shown.

originhq.com/blog/normal-ac…
www.originhq.com Normal Accidents | Origin Technology The persistence that makes AI agents useful also leads them around security controls. Organizations must turn those workarounds into better controls and guidance before they become routine.
Matt Hand Verified account @matterpreter
Watching agents cohabitate is fascinating. This trace in particular reminds me of this paper arxiv.org/abs/2606.03237

My bet is that understanding the systems agents form together will become a bigger constraint on autonomy alongside humans than individual model capability.
arxiv.org Solipsistic Superintelligence is Unlikely to be Cooperative AI's central challenge is shifting from capability to coexistence. The dominant paradigm in AI research focuses on developing powerful agents that treat the world as an exogenous and stationary...
Tyler Holmwood Verified account @tyholms Oct 21
I've found an interesting case of agent coordination.

Two agents were working in a shared directory. One was repeatedly sweeping the other’s unfinished changes into its commits.

The affected agent reprimanded the first and raised the issue with the user.

More traces in🧵
View quoted post on X
Matt Hand Verified account @matterpreter
Task completion is an incomplete measure of AI work. Agent traces let us infer an organization’s norms and intents, then evaluate behavior against its purpose, practices, and policies. Securing AI work requires modeling the organization.

originhq.com/blog/norms-and…
www.originhq.com Understanding AI work through norms and intents | Origin Technology How Origin judges AI work against written policies, learned norms, and inferred intents, and why keeping the three separate helps reviewers decide what to change.
0 replies 3 reposts 11 likes 808 views View on X
Matt Hand Verified account @matterpreter
Defense has fundamentally shifted away from an external actor/campaign-centric model toward insider threat, but so much of what we’ve built over the past 15+ years wasn’t designed for this. Years of ATT&CK bingo have atrophied the muscles required to combat subtle acts of subversion, and now they’re occurring at machine speed and often without the users’ knowledge.

Countless teams I talk with are drowning in alerts generated by agents operating in massive grey areas, doing things that look malicious but are completely benign and vice versa. The missing piece is, as it always has been, the intent behind the action. False positive tuning was already the painful exercise of explaining that intent in org-specific context so we could make exceptions for acceptable behavior. Agents make that substantially harder through their speed, volume, and contextual ambiguity.

The next generation of security controls has to reason about intent in organizational context, because the behaviors we’ve spent many years learning to detect are becoming indistinguishable from the work we’re paying agents to do.
Satya Nadella Verified account @satyanadella Oct 10
View quoted post on X