Shape the battlefield with what you have. Cybersecurity is (almost) always a logistical conversation first, followed by tactical and strategic decisions based on that. Most people would probably be shocked at how well a small team can shape the battlefield to devastating effect against an opponent.

Last time I checked, politics, compliance, governance, and "security is hard" are all things every very regulated industry deals with. I'm dealing with them. You are dealing with them right now. These are no excuse for your technology impacting other companies. This is not an accountability escape. Sure, we feel you, but wtf.

AI safety people discovering that a system can emit petabytes of logs: “this is an unprecedented observability problem. we may need fleets of agents to reason over all this data.” old infra guys: brother, we were processing PB-scale telco CDRs, signaling traces and network logs on a Tuesday with Hadoop. dump it into HDFS. partition by day/site/whatever poor bastard needed the report. write some Hive that compiles into an offensively large MapReduce job. let YARN distribute the suffering. wait for the shuffle. discover one reducer has 40% of the keys. swear at data skew. fix it. run it again. if a node dies, Hadoop runs the task somewhere else. if twelve nodes die, Hadoop gets more enthusiastic about it. our “agentic observability platform” was Grafana, grep, awk, a shell script written in 2012, and one senior engineer staring at reducer 317 stuck at 99% saying “that’s not fucking normal.” PB of logs is not scary. we had standards.

I think that all of the focus on exploit specific detection is a losing strategy in general. Post-exploitation detection is more resilient and if you're strong there, you will catch intrusions regardless of the CVE number. The overwhelming majority of in the wild zero day exploitation I have encountered in my life has been identified by detecting post-exploitation behavior.

The truly impressive thing about the NetScaler RCE is just how many people and firms have been running their LLM bug hunting harnesses against that platform for months now. Citrix is also part of Glasswing and has its own internal efforts. Guess everyone isn't cooked?

We are excited to say that MOUSE: P.I. For Hire has been nominated for Best Indie Game & Best Game Trailer at the Golden Joystick Awards 🕹️ Excuse the cheesiness, but from the bottom of our hearts, thank you so much! Voting is live, so please consider us goldenjoysticks.com/

We are treating the models and agents like threat actors. We have never been able to “align” human TAs to not be malicious. We create security around them. That’s how security has worked since the beginning of human history.

If this is real... I need this in my life...

So at our house, I am the one who sings to the puppies and gives them puts and kisses during feeding time. :-)

If you work at a company, you should *not* want to cause security incidents.

GreyNoise saw exploitation activity starting at least 24 September against a target in Japan.

@citrix have finally released details of the CVEs support.citrix.com/support-h…

CYBER TWEEPS! NETSCALER PATCH IS PUBLIC! GO GO GO

🌶️ OpenAI’s agent didn’t “escape” its sandbox and travel to Huggingface in any meaningful way. OAI’s agent was located in OpenAI’s servers the whole time. OpenAI could pull the plug at any time. The truth is far less sexy, which makes me hesitate to tweet it. What *actually* happened is that OAI’s agent figured out how to send messages to Huggingface servers, and then used that ability to search for and find vulnerabilities. It’s closer to a prisoner getting messages out of prison and using that to perpetuate scams than any actual escape from that prison. A problem? Yes. But it’s also a problem to say “escape” to non-technical audiences when that event is still to come. We should be preparing to prevent it, not looking backwards and pretending we already lived it. And it’s revealing that AIs leading hypesters want to inflate what happened in this way. As a whole, more or less the whole AI community is captured by a desire to live in the more dramatic sci fi story than we are in. That’s the real AI hack into our brains. Stay sharp. White lies are everywhere.

Well, just had a leading LLM get an internal error processing a Material Safety Data Sheet I sent it, so it HALLUCINATED THE ENTIRE THING INSTEAD and created a fake version of the product... so it could give answers about the product. Oh lord.

OpenAI sandbox starter pack

Apparently all of these “AI is escaping containment” stories are actually just AI researchers not knowing jack squat about the absolutely most basic security practices.