AI Security Trend Roundup — Jul 10, 2026

Covering Jul 03 → Jul 10, 2026. 56 new items from 8 tracked sources.

This digest credits every source by name and links directly to each original post. Editorial curation by FixTheVuln — all rights and attribution belong to the original authors.

Academic & Research

• Controllability-Aware Adversarial Examples Against LLM-Based Network Traffic Classifiers

Source: arXiv cs.CR — Jul 10 arXiv:2607.07739v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly explored as network intrusion detection classifiers, but their adversarial robustness under realistic attacker constraints remains unclear. We present a controllability-aware black-box t

• Forensic Schema for Psychological Manipulation in Cyber Fraud: LLM-Driven Victim Reports Analysis

Source: arXiv cs.CR — Jul 10 arXiv:2607.07751v1 Announce Type: new Abstract: Existing cybercrime classification schemas capture contact metadata and financial transactions but omit the psychological manipulation techniques perpetrators employ. We present a forensic schema (four categories, 35 questions) addi

• ScopeJudge: Cost-Aware Pre-Execution Gating for Offensive Security Agents

Source: arXiv cs.CR — Jul 10 arXiv:2607.07774v1 Announce Type: new Abstract: As LLM agents take on offensive security work, a single out-of-scope tool call can breach a client's engagement boundary, disrupt production, or void a bug-bounty finding. Unlike a fixed safety policy, the boundary that matters is d

• Open Models, Open Risks: Measuring Unsafe Generation in Text-to-Image Models In the Wild

Source: arXiv cs.CR — Jul 10 arXiv:2607.07827v1 Announce Type: new Abstract: Existing safety studies on text-to-image (T2I) jailbreaks are largely conducted in controlled in-the-lab settings, typically on a small number of canonical models. As a result, the current safety status of the rapidly growing in-the

• Mechanistic Interpretability of LLM Jailbreaks via Internal Attribution Graphs

Source: arXiv cs.CR — Jul 10 arXiv:2607.07903v1 Announce Type: new Abstract: Large language models (LLMs) exhibit remarkable capabilities but remain highly vulnerable to adversarial prompts and jailbreak attacks. Existing approaches primarily analyze these failures through input-output behaviors or attributi

• Who Broke the System? Failure Localization in LLM-Based Multi-Agent Systems

Source: arXiv cs.CR — Jul 10 arXiv:2607.07989v1 Announce Type: new Abstract: Large language model (LLM) based multi-agent systems enable complex problem solving through coordinated reasoning and action, but their distributed structure also introduces new challenges in diagnosing system-level failures. When a

• zkComposer: Decomposing Proof Construction to Scale zkML

Source: arXiv cs.CR — Jul 10 arXiv:2607.08095v1 Announce Type: new Abstract: Zero-knowledge machine learning (zkML) enables a server to perform verifiable inference while keeping model parameters private from the client. However, existing zkML systems incur prohibitive proof-generation costs. We observe that

• Prismata: Confining Cross-Site Prompt Injection in Web Agents

Source: arXiv cs.CR — Jul 10 arXiv:2607.08147v1 Announce Type: new Abstract: Autonomous web agents promise to automate everyday browsing tasks, but inherit one of the web's oldest attack surfaces. Cross-Site Scripting proved that mixing trusted and untrusted content is dangerous, even on benign pages. Agents

• Out of Sight: Compression-Aware Content Protection against Agentic Crawlers

Source: arXiv cs.CR — Jul 10 arXiv:2607.08180v1 Announce Type: new Abstract: The rise of LLM-based agents with reasoning, summarization, and memory capabilities has created a new threat surface for online content that conventional defenses fail to address. Existing defenses like access controls can be circum

• MLQENABLER: Enabling Secure Machine Learning Queries over Encrypted Database in Cloud Computing

Source: arXiv cs.CR — Jul 10 arXiv:2607.08197v1 Announce Type: new Abstract: In cloud computing, the public cloud service providers (CSPs) can provide cloud storage as the primary service while providing additional machine learning (ML)-based services by using the clients' data in storage. This business mode

• Simulating the Resident: Generating Executable Smart Home Schedules via LLM Personas

Source: arXiv cs.CR — Jul 10 arXiv:2607.08231v1 Announce Type: new Abstract: Smart homes have emerged as an important domain for HCI research, including work on usable security and privacy. Ideally, studies in these areas draw on datasets collected in real homes with real residents, capturing authentic devic

• Multi-Agent Firewall Architecture for Privacy Protection of Sensitive Data in Interactions with Language Models

Source: arXiv cs.CR — Jul 10 arXiv:2607.08282v1 Announce Type: new Abstract: While Large Language Models (LLMs) have become essential productivity tools, their integration into workflows without adequate safeguards creates significant risks. This paper proposes an open-source, privacy-focused, user-facing fi

• From Legacy Documentation to OSCAL: An MCP-Based Agent Pipeline for Threat-Informed Continuous Compliance in Critical Infrastructure

Source: arXiv cs.CR — Jul 10 arXiv:2607.08288v1 Announce Type: new Abstract: In critical infrastructure, operational technology environments often cannot be actively scanned, and yet active system feedback is needed for risk assessment and compliance. This paper presents a non-invasive, MCP-grounded multi-ag

• TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories

Source: arXiv cs.CR — Jul 10 arXiv:2607.08400v1 Announce Type: new Abstract: LLM agents reach users through resellers, who may rebrand a developer's agent or substitute a cheaper model. When provenance is disputed, attribution rests on the trajectory log (the record of tool calls, observations, and executed

Prompt Injection & LLM Security

• Quoting OpenAI

Source: Simon Willison — Jul 10 [...] Work on web and mobile runs in the cloud. Work in the desktop app can also use local files and desktop apps with your permission. At launch, cloud Work conversations do not appear in desktop Work; desktop Work threads and local files remain on that computer. — OpenAI,

• The new GPT-5.6 family: Luna, Terra, Sol

Source: Simon Willison — Jul 09 OpenAI's latest flagship model hit general availability this morning, and comes in three sizes: Luna, Terra, and Sol (from smallest to largest). The new models are priced per 1M input/output tokens as Luna $1/$6, Terra $2.50/$15, Sol $5/$30. For comparison, the Claude Opus series

• Introducing Muse Spark 1.1

Source: Simon Willison — Jul 09 Introducing Muse Spark 1.1 Following Muse Spark in April, here's Muse Spark 1.1 - the first Spark model to offer an API. Meta claim significant improvements in agentic tool calling and computer use. There are a lot more details are in the Muse Spark 1.1 Evaluation Report. The "At

• llm-meta-ai 0.1

Source: Simon Willison — Jul 09 Release: llm-meta-ai 0.1 Let's LLM run prompts against the new muse-spark-1.1 model. Tags: llm, meta

• llm 0.31.1

Source: Simon Willison — Jul 09 Release: llm 0.31.1 Fix for a bug with OpenAI Chat Completion endpoints where a tool call with empty arguments could result in a JSON error from some providers. #1521 This bug came up when I was testing llm-meta-ai. Tags: llm

• Rewriting Bun in Rust

Source: Simon Willison — Jul 08 Rewriting Bun in Rust Jarred Sumner has been promising this blog post (since May 9th) about his Zig to Rust rewrite of Bun for significantly longer than it took him to finish the rewrite. Honestly, it was worth the wait. This is a detailed description of an extremely sophisticate

• Introducing GPT‑Live

Source: Simon Willison — Jul 08 Introducing GPT‑Live OpenAI finally upgraded the model used by ChatGPT voice mode! I've had preview access for a few weeks in the iPhone app, and the new model is very impressive. It also has the ability to spin off harder tasks to GPT-5.5: For questions that require web search,

• Quoting Kenton Varda

Source: Simon Willison — Jul 08 I just declared a moratorium against AI-written change descriptions (e.g. PR and commit messages, also issues/tickets) from my team. AI was writing change descriptions that were worse than useless to me as I tried to review PRs: outlining details of the code that could easily be

• sqlite-utils 4.0, now with database schema migrations

Source: Simon Willison — Jul 07 This morning I released sqlite-utils 4.0, the 124th release of that project and the first major version bump since 3.0 in November 2020. In addition to some small but significant breaking changes (described in this upgrade guide), this version introduces three major features: dat

• sqlite-migrate 0.2

Source: Simon Willison — Jul 07 Release: sqlite-migrate 0.2 The version that retires the library, instead implementing a compatibility shim against the new sqlite-utils 4.0 dependency. Tags: sqlite-utils

• github-code Web Component

Source: Simon Willison — Jul 07 Tool: github-code Web Component An experimental Web Component built using GPT-5.5 and the following prompt: let's build a Web Component for embedding code from GitHub sqlite-utils 4.0

Source: Simon Willison — Jul 07 Release: sqlite-utils 4.0 See sqlite-utils 4.0, now with database schema migrations for details. Tags: sqlite-utils

• sqlite-utils 4.0rc4

Source: Simon Willison — Jul 07 Release: sqlite-utils 4.0rc4 The last RC before the 4.0 stable release. Mainly implements feedback from a detailed review by Claude Fable 5. Tags: sqlite-utils, claude-mythos-fable

• tencent/Hy3

Source: Simon Willison — Jul 06 tencent/Hy3 New Apache 2.0 licensed model from Tencent in China: Hy3 is a 295B-parameter Mixture-of-Experts (MoE) model with 21B active parameters and 3.8B MTP layer parameters, developed by the Tencent Hy Team. Following the Hy3 Preview launch in late April, we gathered feedback

• sqlite-utils 4.0rc3

Source: Simon Willison — Jul 06 Release: sqlite-utils 4.0rc3 I hoped to release sqlite-utils 4.0 stable this weekend, but as I worked through the backlog of issues and PRs with a combination of Claude Fable 5 and GPT-5.5 the changelog since rc2 kept getting bigger. The biggest new feature is support for introsp

• sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25)

Source: Simon Willison — Jul 05 I wrote about the sqlite-utils 4.0rc1 release a couple of weeks ago. Since we only have Claude Fable on our Max subscriptions for a few more days, I decided to see if it could help me get to a 4.0 stable release that I felt truly comfortable about, since I try to keep to SemVer a

• sqlite-utils 4.0rc2

Source: Simon Willison — Jul 05 Release: sqlite-utils 4.0rc2 See sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25).

• Building a World Map with only 500 bytes

Source: Simon Willison — Jul 04 Building a World Map with only 500 bytes Iwo Kadziela (assisted by Codex) figured out a way to generate a credible ASCII world map using 445 bytes of data: The key trick is to use deflate compression, which is then wired together using this neat snippet of JavaScript. I didn't kn

• Better Models: Worse Tools

Source: Simon Willison — Jul 04 Better Models: Worse Tools Armin reports on a weird problem he ran into while hacking on Pi: The short version is that newer Claude models sometimes call Pi’s edit tool with extra, invented fields in the nested edits[] array. And not Haiku or some small model: Opus 4.8. The edit

• Open Source AI Gap Map

Source: Simon Willison — Jul 03 Open Source AI Gap Map Current AI is "a global partnership building a public option for AI", founded as a non-profit at the AI Action Summit in Paris in February 2025 and backed by serious capital ($400m already committed). They launched their Gap Map a couple of days ago - an at

• Quoting Josh W. Comeau

Source: Simon Willison — Jul 03 I just launched my third course, Whimsical Animations, and so far, it’s on track to sell roughly ⅓ as many copies as a typical course launch. It’s a similar story with my two existing courses. Sales are down significantly from last year. There are likely a lot of reasons for this

• Fable's judgement

Source: Simon Willison — Jul 03 One of the most interesting tips I got from the Fireside Chat I hosted with Cat Wu and Thariq Shihipar from the Claude Code team at AIE on Wednesday was to let Fable (and to a certain extent Opus) use their own judgement rather than dictating how they should work. The example the

Community Signal

• How do you use Vim in the era of AI?

Source: Hacker News (AI Security) — Jul 10 I'm a heavy vim user, but now that I look at the code less and less I'm wondering if it still makes sense. Comments URL: https://news.ycombinator.com/item?id=48859439 Points: 28 # Comments: 60

• AI doesn't know how to forgive and cannot forget

Source: Hacker News (AI Security) — Jul 10 Article URL: https://tejassuds.com/blog/ai-cannot-forget Comments URL: https://news.ycombinator.com/item?id=48859047 Points: 21 # Comments: 21

• Ask HN: Do we need a support group for developers alienated by LLMs?

Source: Hacker News (AI Security) — Jul 10 Surely I must not be the only one developer that doesn't recognize this field any more, and would rather do anything else than become a glorified code reviewer for machine generated code.Seeing my career crumble in front of my eyes, seeing my identity as software engineer questio

• AI-generated videos to maximally drive a target brain region

Source: Hacker News (AI Security) — Jul 10 Article URL: https://nevo-project.epfl.ch/ Comments URL: https://news.ycombinator.com/item?id=48856904 Points: 184 # Comments: 187

• China may restrict foreign access to Chinese open-source AI models

Source: Hacker News (AI Security) — Jul 10 Article URL: https://www.reuters.com/technology/artificial-intelligence/china-weighs-silicon-curtain-around-sought-after-ai-models-2026-07-08/ Comments URL: https://news.ycombinator.com/item?id=48856412 Points: 37 # Comments: 0

• Build your own vulnerability harness

Source: Hacker News (AI Security) — Jul 10 Article URL: https://blog.cloudflare.com/build-your-own-vulnerability-harness/ Comments URL: https://news.ycombinator.com/item?id=48854681 Points: 60 # Comments: 21

• Building a real-time AI tutor for 5-year-olds

Source: Hacker News (AI Security) — Jul 09 Hey HN! We've spent the good part of this past year building an AI tutor that teaches kids ages 4-9 reading, math, ESL and more. Getting an AI tutor to effectively teach a child turns out to be a really hard technical challenge, this took getting the underlying architecture right

• GLM 5.2 is nearly as accurate as a human book keeper

Source: Hacker News (AI Security) — Jul 09 Article URL: https://toot-books.pages.dev/blog/glm-5-2-vat-benchmark Comments URL: https://news.ycombinator.com/item?id=48850414 Points: 217 # Comments: 117

• Show HN: I built a web tool to see and edit what an AI thinks before it answers

Source: Hacker News (AI Security) — Jul 09 I run a small AI lab and playground and got super excited about Anthropics paper "Verbalizable Representations Form a Global Workspace in Language Models" (https://transformer-circuits.pub/2026/workspace/index.html)It talks about how they use a tool they call a Jacobian Lens to v

• AI 2040: Plan A

Source: Hacker News (AI Security) — Jul 09 Article URL: https://ai-2040.com/ Comments URL: https://news.ycombinator.com/item?id=48848425 Points: 85 # Comments: 49

• AI content is everywhere on social media, especially LinkedIn

Source: Hacker News (AI Security) — Jul 09 Article URL: https://www.pangram.com/blog/ai-in-your-feed Comments URL: https://news.ycombinator.com/item?id=48847940 Points: 229 # Comments: 206

• The next era of AI is about infrastructure, not just models

Source: Hacker News (AI Security) — Jul 09 Article URL: https://blog.mozilla.ai/the-control-layer-why-the-next-era-of-ai-is-about-infrastructure-not-just-models/ Comments URL: https://news.ycombinator.com/item?id=48846842 Points: 55 # Comments: 23

• DeepSeek aims to make its own AI chip

Source: Hacker News (AI Security) — Jul 09 Article URL: https://www.proactiveinvestors.com/companies/news/1095178/deepseek-makes-pivot-that-should-put-silicon-valley-on-high-alert-1095178.html Comments URL: https://news.ycombinator.com/item?id=48846708 Points: 66 # Comments: 8

• Show HN: FableCut – A browser video editor AI agents can drive (zero deps)

Source: Hacker News (AI Security) — Jul 09 Article URL: https://github.com/ronak-create/FableCut Comments URL: https://news.ycombinator.com/item?id=48845422 Points: 94 # Comments: 58

• Show HN: Getting GLM 5.2 running on my slow computer

Source: Hacker News (AI Security) — Jul 09 A few days ago I found myself trying out GLM 5.2 and was really positively impressed. The capabilities and security I was getting from this LLM are similar to those I've gotten from models like Claude or GPT, and this really surprised me.But then I thought, "I wonder how it would

• AI changes the economics of software rewrites

Source: Hacker News (AI Security) — Jul 09 Article URL: https://thetruthasiseeitnow.com/ai-slop-starts-with-the-codebase-itself/ Comments URL: https://news.ycombinator.com/item?id=48841446 Points: 102 # Comments: 106

• AI software that generates 'rage bait' developed by Germany's far-right AfD

Source: Hacker News (AI Security) — Jul 09 Article URL: https://www.irishtimes.com/world/europe/2026/07/08/ai-software-that-generates-rage-bait-developed-by-germanys-far-right-afd/ Comments URL: https://news.ycombinator.com/item?id=48841275 Points: 23 # Comments: 4

• What's slowing down the AI buildout

Source: Hacker News (AI Security) — Jul 09 Article URL: https://www.worksinprogress.news/p/ai-is-bottlenecked-by-the-grid Comments URL: https://news.ycombinator.com/item?id=48840620 Points: 78 # Comments: 185

• I think I have LLM burnout

Source: Hacker News (AI Security) — Jul 09 Article URL: https://www.alecscollon.com/blog/llm-burnout/ Comments URL: https://news.ycombinator.com/item?id=48839984 Points: 396 # Comments: 349

• Suspecting AI cheating, Ivy League prof ordered in-person final; scores fell 50%

Source: Hacker News (AI Security) — Jul 08 Article URL: https://arstechnica.com/ai/2026/07/we-cannot-choose-to-become-idiots-the-ai-cheating-scandal-roiling-brown-university/ Comments URL: https://news.ycombinator.com/item?id=48838611 Points: 133 # Comments: 156

---

Source List

All sources tracked in this roundup, credited to their original authors/organizations:

• OWASP GenAI Security Project — feed: https://genai.owasp.org/feed/
• Simon Willison — feed: https://simonwillison.net/atom/everything/
• arXiv cs.CR — feed: http://export.arxiv.org/rss/cs.CR
• Protect AI — feed: https://protectai.com/blog/rss.xml
• Google Project Zero — feed: https://googleprojectzero.blogspot.com/feeds/posts/default
• CISA Cybersecurity Advisories — feed: https://www.cisa.gov/cybersecurity-advisories/all.xml
• NIST Cybersecurity News — feed: https://www.nist.gov/news-events/cybersecurity/rss.xml
• Hacker News (AI Security) — feed: https://hnrss.org/newest?q=%22AI+security%22+OR+%22prompt+injection%22+OR+%22LLM+vulnerability%22&points=20