Uncategorized

BIML is proud to host Patrick McDaniel, an OG of machine learning security (prominently featured in the BIML TOP 5) and a Dean of Research at Wisconsin, for a visit to the BIML Barn. Patrick arrived in Berryville late on Thursday and was greeted with a Liberal or two on the porch. We stayed up way too late talking about AI and security.

In the morning after breakfast, we spent much of the Friday research discussion going over our soon to be released paper No Security Meter for AI. Patrick has been thinking about measuring ML behavior for a long time, and was an early proponent of a whitebox approach. He had lots of very useful feedback for us.

Does science really get done around the kitchen table? Why yes. Yes it does. (And technical talks really get delivered in the BIML Barn.)

We ventured into greater metropolitan Berryville for lunch and coffee.

And then Patrick delivered a new talk as a BIML in the Barn feature to be released on May 13th. Patrick’s talk really surprised us and in very important philosophical ways.

After the talk we shared a cocktail on the patio. Maybelline is an honorary BIML dog.

Patrick enjoys a well-deserved Lemon Mint Fizz.

And then it was off to dinner with BIML spouses at Huntōn in Leesburg.

Fantastic visit. These kinds of human interaction are absolutely critical as we construct a reasonable approach to machine learning security.

Gary McGraw, cofounder of the Berryville Institute of Machine Learning, pointed to a core gap: Today’s benchmarks tend to measure how well AI systems can perform security tasks—not how secure the systems themselves are. Companies need to keep that distinction in mind when evaluating their tools and defenses.

McGraw warned as far back as 2019 that securing machine learning systems would be “one of the defining cybersecurity struggles of the next decade.” That moment has now arrived.

“These meetings are a way to remind ourselves of the fundamentals,” he said, “as we try to define what machine learning security actually is.”

What was to be a more standard copy of the BIML risk talk, instead was transformed into a debut of BIML’s forthcoming paper No Security Meter for AI. (expected mid-May) for an audience of NIST computer scientists.

It’s always fun to debut a talk for an audience that is engaged and knowledgeable.

While we were inside the very industrial Chemistry building for a talk that was 80% zoom, it rained outside.

Well, maybe. (McGraw proposed the name which is being vetted.) We did all get together in Arlington 4.21.26 to discuss policy and AI. It was a good meeting set up by OWASP and SANS and run very professionally by Rob van der Veer.

The cool thing? BIML’s work was not only cited, but included.

The meeting setting was gorgeous.

As usual, the hall track was the best part of the entire day…especially when the hall was moved across the street to the bar.

See this coverage of the meeting: Global AI Security Standard Organizations Gather Under MOSAIC to Reduce Fragmentation, AI security leaders gather in Washington as risks mount—and Mythos raises the stakes

Decipher covered the controversy over Mythos and Glasswing in an insightful article (which featured BIML’s opinion prominently). Have a look at the article and the video attached below.

Apparently, AI was used to steal original art from an aspiring musician. This is something that eminent philosopher Dan Dennett warned us about in 2022 (see the video link at the bottom of the BIML Bibliography). Though we mostly focus on technical risks built in to ML, the risk of deep fakes is a very real one that deserves more attention.

Have you heard? The mythos model from Anthropic is so dangerously good at finding software vulnerabilities that its release must be initially limited to companies participating in the Glasswing software security project! {Oh my. Also lions and tigers and bears!}

Does that sound like a marketing ploy to you? Because it does to most expert bug finders that I know best. In fact, the software exploit community (some of whom make a very good living selling bugs to the very companies that produced them…LOL) is pretty evenly split on this issue. So what is a grownup to think?

Those of who have been around the block a few times in AI-land remember way back when Chat-GPT2 was too dangerous to release too (because it could generate fake news even faster than a political PR flak). That garnered some press and helped with the launch for sure. Well, it’s happening again…just look at the tech headlines! Go, Anthropic, go!

Fortunately, there is some balanced coverage out there adopting a thoughtful approach (thanks, Cade). Here’s what we think:

1. We still have a very real software security problem, so ANYTHING that helps people find AND FIX bugs in code is good. Everyone who is serious about software vulnerability has been using Agentic AI to do this better. You should too. Want to get started using AI to find bugs? Hold your nose (because LinkedIn) and check out this link. But please also figure out how to FIX the bugs you find. And don’t expect to be paid for slop.
2. LLMs really are good at helping find easy vulnerabilities, but expert mode requires human experience and expertise. Will you become Halvar Flake by strapping on mythos? No, you will not.
3. Building exploits that really work is much harder than just finding bugs. In fact, I wrote a whole book about this in 2004, 22 years ago, and it is still true. Patching is also harder than finding vulnerabilities. Hopefully AI will help with both of these software security activities.
4. AI tools are all helpful in different ways. Use them all. Use the ones that are already released. (We hear tell that a well prompted Opus-4.6 (82%) does nearly as well as Mythos (84%) on CRSBench…which calls into question just what the hell these benchmarks measure—a topic we have been thinking about a bunch.)

As a last thought, we’re going to appeal to the four I’s that excellent human designers are familiar with: Intuition, Insight, and Inspiration (the fourth one is the “self” kind of I). AI is great and we love it. We are really going to need lots more software architects, information architects, designers, actual building architects, and humans who know what they are doing. If you know what you’re doing, you’ll be fine. If you are simply a bullshitter, you’re toast.

One of our key missions at BIML is to help establish the field of machine learning security. Towards that end, we welcome collaboration with academics and practitioners alike. The AI Cyber Lab straddles both targets at once.

We held our first meeting with Neil Daswani’s AI Cyber Lab March 26th. Neil is a well-known figure at the intersection of academic security (having taught many classes at Stanford where he earned his Ph.D.) and applied security (serving as CISO at Lifelock after time spent in applied security at Yodlee and Google). These two plus decades of experience inform current innovation in cybersecurity and AI. (You may also know Neil as the author of Big Breaches: Cybersecurity Lessons for Everyone).

The AI Cyber Lab team encompasses seasoned CISOs and machine learning architects as well as college students focused on agentic AI and security. In this initial meeting, we introduced what BIML is focused on, defining Machine Learning Security and our unique approach. Gary delivered a quick informal presentation presentation.

We look forward to future collaboration and sharing research findings as we hack our way through the MLsec jungle.

Imagine that you are trying to practice good security engineering at the system level when one of your essential components is an unpredicatable black box that sometimes does the wrong thing. How do you ensure or even measure the trustworthiness of that system? That seems to be the current situation we are in with LLMs and Agentic AI.

One of the levers we are exploring is observability INSIDE the black box. SO, In the case of an LLM, that would be trying to figure out what is going on inside the Transformer. Are there circuits in the trained model that correlate with and define certain behaviors? Are there concepts in there? Can we make use of various activation patterns (and weights) or otherwise guide them from inside the network? Are there indicators of bad behavior? Can we see the “guidelines” imposed by alignment training? Are they robust? Etc.

This is what we call (for the moment anyway) “Whitebox Interpositioning” at BIML. It’s like watching your brain (and interposing inside it) while you are acting as part of a system. Maybe we can build an “Intention-ometer” or maybe not. But we are certainly moving toward “WHYness” in a WHAT machine.

This all reminds us of what happened in software security when we moved from black box monitoring and sandboxing to whitebox code analysis (static and dynamic both). Thing is, we never really got a handle on architecture, especially when it came to security…

Plenty of work to do on the raw science front…and something we want to create a coalition to approach. Toward that end, BIML recently hosted a whitebox summit with Realm Labs and Starseer. We were joined by Paul Kocher. Expect something to come of this.

One of our key missions at BIML is to define the future of machine learning security. [un]prompted was hugely helpful in that regard, and we are proud to have participated.

All in one place; real people leading important work in MLsec.

The [un]prompted conference delivered. No frills, all substance. This is where AI researchers and security practitioners met to share what they are seeing and doing across the new world of machine learning security and AI vulnerability risk.

Anthropic’s Nicolas Carlini delivered an excellent talk titled “Black-hat LLMs,” all about automating attack with AI tools. The urgency came through—we are at a very real inflection point. Carlini implored the audience to “help make the future go well!!!” (by being part of the solution making #AI as secure a possibe…) in a room packed with peers from OpenAI, Google Deepmind, Nvidia, Salesforce, founders of early stage AI companies, and actual real life hackers and security engineers.

(*) Carlini features in BIML’s TOP 5 (our research group curates an extensive annotated bibliography here) for his work on Data Extraction.

(*) Another star in the field, Ilia Shumilova whose 2023 paper on Recursive Pollution is also in our Top 5was in townrepresenting his start-up Sequrity AI.

(*) Carl Hurd of Starseer shared how his startup is revolutionizing MLsec by opening the black box, and looking inside to see what is actually going on. (See a posting by Carl about his talk here.)

In all, the conference was packed with two tracks of speakers selected from over 500 submitted proposals. Thank you to everyone who submitted talks. And a massive thank you to the sponsors KnosticTachTechAISLE, Whiterabbit, Halcyon Futures, Halcyon Ventures and for the hard work of Gadi, Kyle, Pedram, Ida, Sounil and many others.

And, one more thing…. you can engage in the content of the conference via this [un]prompted 2026 NotebookLM creation by Rob T. Lee – amazing!