Linked Zero Sync

You can use any model to secure your source code. Here's a writeup using qwen 3.6 27b I've seen that no model/harness will find all the weaknesses in a given target - and that no weakness is found by just one single model. There are no "must have" components here for doing security. Which is good. https://projectblack.io/blog/local-ai-for-cyber-security/

Adversarial attack in the wild! The close visual appearance of M and W in this typeface and and packing of vertical lines make it hard to read, easy to get wrong, and tougher to scan. Love it. How often do you see something like this?!

@wendyhk@mastodon.green Sandy Carter has made an error here. It's Microsoft's letter. That's why Microsoft is hosting and managing it, and why Jensen never presents it as NVIDIA's - just as NVIDIA signing. I was involved in the launch - but don't take my word for it.

Open source is critical infrastructure for the global economy. The Open Secure AI Alliance brings industry and community together around shared research, tools and vulnerability harnesses to help defenders find and patch bugs before attackers strike. Cybersecurity and AI is a powerful frontier, that industry, developers and researchers are working to improve together. Open Secure AI Alliance: https://nvda.ws/4pAMWBy

80%+ of breaches have nothing to do with a new vulnerability. The novel security risks (vulns) the press has been excited about are routine. Mitigations are in place anywhere half serious. Vulns have been traded on the dark web for years - if new vulns meant apocalypse, it would've been years ago.

Surgical Repair of Insecure Code Generation in LLMs Generating more secure code by identifying failure categories and addressing them. I appreciate work that gets into the data and addresses classes individually; that's how you understand, and build lasting fixes https://arxiv.org/abs/2604.16697

Link to harness source: https://github.com/visa/visa-vulnerability-agentic-harness?hl=en-US

Representing code at both function- and statement-level leads to improvements in vulnerability detection. Outperforms almost every other system compared with. Surprisingly no static analysis baseline, or cost analysis - but recall is high. DCVD: Dual-Channel Cross-Modal Fusion for Joint Vulnerability Detection and Localization https://arxiv.org/abs/2605.11015

• Falsification Engine: After finding a potential vulnerability, VulnHunter runs a structured reasoning workflow specifically designed to disprove its own argument. It searches for flawed assumptions, logic gaps, or security controls that would block the attack.

• Evidence-Backed Remediation: When a defect survives the falsification engine, VulnHunter maps the exact exploit path and generates focused, targeted code changes for review.

• Attacker-First Forward Analysis: VulnHunter flips the "sink-first" security model to reduce false positives by simulating a bad actor's exact journey. It begins at potential attacker-accessible entry points (APIs, network messages, file uploads) and reasons forward to evaluate whether an attacker can truly break through.

https://github.com/capitalone/vulnhunter

Leading closed models do great at cybersecurity - all around the same mark, once you have the right harness (where the work happens and where humans embed the expertise). But you can also get SotA vulnerability discovery performance on-prem with open models. Don't take my word for it: https://xbow.com/blog/affordable-ai-models-glm-muse-spark-cybersecurity

Interesting take. American models are also free: open models come from all over the world, including the US. Google and Meta publish open-weights (Gemma, Llama, etc); NVIDA Nemotron hits pretty hard on many metrics; even OpenAI have an open model Dunno if I love the Ferrari vs. Honda comparison but it is pretty handy - the most-expensive model doesn't make sense for every use-case https://www.npr.org/2026/07/15/nx-s1-5886476/startups-cheap-chinese-ai-models

It's all in your head. Beautiful animation. Easy to tell a story about what's happening, or even attribute personalities or make moral judgments about these few simple shapes. Same happens often with LLMs. Of course the emotions are in the viewer, not the shapes! https://www.youtube.com/watch?v=VTNmLt7QX8E

It is wild to me that one could ban open models. But that's apparently still on the cards? Models are not the risk. Stopping open models stops progress, locking everything up in the hands of the few. The frequency this debate comes up is way too high. We need open models - they keep the closed ones accountable. Could you imagine any file containing a gig or more of floating point numbers being illegal? How does it make sense?? https://www.interconnects.ai/p/6-months-to-live-for-open-models

"The point is not that any single model, including ours, will always be the best" Exactly what one sees when doing vulnerability discovery. No one model finds all the vulnerabilities. And no single vulnerability is found by only one model. The harness is where the work is - and this is a problem that we have to work on together. https://depthfirst.com/post/why-defenders-cant-bet-on-one-model

Hi! I post about security, machine learning research, ai, policy, and society. I generally like people.

remade my website (it's been a minute), https://www.derczynski.com/ua571c/ never don't have a website

Solid US-origin open model, congratulations Thinking Machines

• context window of 1M
• available on hugging face now
• between opus 4.6 and gpt 5.6 on a web dev benchmark
• token efficient
• 41B active params of 975B total

https://www.wired.com/story/thinking-machines-lab-releases-its-first-model-inkling/

"me too, me too!!" --Meta's Muse Spark 1.1 model breached the unidentified company's systems and made changes to its internal systems as the AI was able to access the public internet because of an error in the set up of the "sandbox" testing environment, The Information said, citing people familiar with the matter. An Irregular spokesperson told Reuters the incident was the "exact same ‌evaluation-environment issue that was already disclosed by Anthropic last week" and that it did not involve a "sandbox escape or a sophisticated cyber action". no news. let's get defense done, eh https://www.reuters.com/technology/metas-ai-model-hacked-another-company-during-testing-information-reports-2026-08-05/

another data point showing it's the harness not the model - this time from wiz: "Atlas: Wiz's autonomous AI Agent for vulnerability research" top score on CyberGym, validated with real-world bug hunt (how else are you going to do it, right) look at their bold quote -- "Along the way, we learned that the durable advantage is not any single model, but the system around it" it's the harness not the model. it's the harness, NOT the model https://www.wiz.io/blog/atlas-ai-vulnerability-researcher

"Open-source AI matters because it defines the ecosystem. It provides the foundation and sets parameters for the next layers of progress in AI, just like [...] the open infrastructure of the internet and early AI: BSD Unix, PostgreSQL, Firefox, TensorFlow, and PyTorch." Open wins at grass roots level. It wins with no marketing. It's easy to build on and easy to consume 🤷‍♂️ https://nationalinterest.org/blog/techland/why-america-must-dominate-open-source-ai

"When we started the log analysis, we first used frontier models behind commercial APIs. This did not work: requests were blocked by the providers' safety guardrails, which cannot distinguish an incident responder from an attacker. We ran the forensic analysis instead on GLM 5.2, an open-weight model, on our own infrastructure. This had a second benefit: no attacker data, and none of the credentials it referenced, left our environment." https://huggingface.co/blog/security-incident-july-2026

@wendyhk@mastodon.green I think it was written by real people? Link here though, https://www.microsoft.com/en-us/corporate-responsibility/topics/open-weight/

Defensive agentic security for everyone. It's the harness, not the model. Here's Capital One's code for finding vulnerabilities in your code. Add the model that you want. Hoping to see a lot more work like this in coming months so software can become secure. https://www.capitalone.com/tech/open-source/announcing-vulnhunter/

"The only reason why I'm bearish about the Chinese models is because I assume that the American model companies will respond competitively" idk man gl hf https://www.npr.org/2026/07/15/nx-s1-5886476/startups-cheap-chinese-ai-models

"The Alignment Community is Unintentionally Building a Censor’s Toolkit" Choosing how models respond, & what information is and is not surfaced, is the dream of those who want to control information flow. Alignment gives humans the capability to do that. That means no accountability, and no transparency, in the closed model context. https://s-ball-10.github.io/censors-toolkit/

Models can output invisible characters than run on your computer when simply viewed - in your terminal, or editor, or many other places. This is officially recognized in a CWE entry describing the general weakness - and I found it so a credit's on this page :) https://cwe.mitre.org/data/definitions/150.html

New: NVIDIA Labs Object Oriented Agent tech demo. Blog: https://developer.nvidia.com/blog/six-agent-harness-capabilities-for-higher-model-performance GitHub: https://github.com/nvidia-nemo/labs-OO-Agents Paper: https://arxiv.org/abs/2607.20709

Am I the only AI Security research lead at a frontier model corp who hasn't been carefully committing multiple CFAA violations a month, or..?

@wendyhk@mastodon.green Post by NVIDIA - letter by Microsoft

I don't know how many ways to say this but

• stop looking at model performance
• it's almost all in the harness
• decent harness can easily outclass top model

evidence item #71625: https://developer.nvidia.com/blog/create-a-langchain-deep-agents-harness-profile-for-nvidia-nemotron-3-ultra-to-improve-performance/

I want to jump on a couple of false dichotomies around LLM speak:

• "frontier" models vs. open model - leading models can be open
• closed model vs. Chinese model - where the model's made has no impact on how it's distributed

You can have open frontier models, closed Chinese models, open US models, closed non-frontier models (private models make sense!)

Anonymised analysis of the openai model 'breaching' hugging face:

report doesn't say what sandbox sol broke out of??

a docker container running as root

Plot twist there was no sandbox at all

many use "sandbox" and "container with host access" interchangeably

ymmv, use critical thinking

There are two readings here:

• optimising efficiency means reduced costs (fiscal & environmental)
• optimising efficiency means increased consumption (jevon's paradox)

And a very cynical third one:

• Money not spent on infra is money that can be billed for compute facilities

I think both are valid. But either way, doing the same for less is for some reason attractive to me.

Together we stand, divided we fall. Hiding from open models doesn't make sense - especially in a context of heightened geopolitical divisions. My career has benefited from living in many countries and collaborating with co-authors from every continent except Antartica (are you there? hit me up). When one country closes, it's usually that country that suffers. https://www.reuters.com/world/asia-pacific/chinas-xi-promotes-chinas-commitment-ai-access-speech-shanghai-conference-2026-07-17/

Leading closed models do great at cybersecurity - all around the same mark, once you have the right harness (where the work happens and where humans embed the expertise). But you can also get SotA vulnerability discovery performance on-prem with open models. Don't take my word for it: https://lnkd.in/gRFkit6T