AI Security Trend Roundup — Aug 14, 2026

Covering Aug 07 → Aug 14, 2026. 52 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

• The energetic cost of mitigating AI attacks in cellular networks

Source: arXiv cs.CR — Aug 14 arXiv:2608.12431v1 Announce Type: new Abstract: The integration of Artificial Intelligence (AI), generally as Machine Learning (ML) algorithms, in all levels and aspects of cellular networks demonstrates the success of data-driven algorithms; for example, the Radio Intelligence C

• Tracing Provenance and Detecting Tampering with Complementary LLM Watermarks

Source: arXiv cs.CR — Aug 14 arXiv:2608.12713v1 Announce Type: new Abstract: Watermarking LLM-generated text is an important task for tracing its provenance. Existing LLM watermarks preserve provenance under editing, but this same robustness allows an adversary to alter critical content while retaining attri

• PIPES: Securing Agent Perception with Provenance and Priors

Source: arXiv cs.CR — Aug 14 arXiv:2608.12789v1 Announce Type: new Abstract: Tool-using agents consume external data from sources with different levels of trust, yet tool responses rarely identify who produced each component or what it should convey. We show that this gap enables state-corruption attacks, in

• Beyond Source: An Empirical Study of Python Bytecode Security Risks

Source: arXiv cs.CR — Aug 14 arXiv:2608.12853v1 Announce Type: new Abstract: Python package security is largely source-centric, yet Python runtimes can execute bytecode directly through .pyc files, compiled-only modules, and marshalled code objects, creating an inspection-execution gap. We present an empiric

• Labels Are Not Endpoints: Treatment Leakage and Construct Validity in MCP Agent Security Evaluation

Source: arXiv cs.CR — Aug 14 arXiv:2608.12880v1 Announce Type: new Abstract: Security evaluations of tool-using agents often equate stored labels with behavioral facts. We audit a preserved campaign by tracing 10,200 execution rows to 180 model-bound requests, 45 semantic requests, and 15 observable stimuli.

• Adversarial Robustness in Smishing Detection: A Comparative Analysis of Adversarial Fragility in Classical vs. Transformer-Based Detection Systems

Source: arXiv cs.CR — Aug 14 arXiv:2608.12889v1 Announce Type: new Abstract: Smishing detection systems are commonly trained and evaluated on clean, monolingual text. In low-resource settings, however, attackers frequently circumvent these systems through character obfuscation, cross-lingual code-switching,

• Beyond Handcrafted Security: Towards Self-Evolving Defense for LLM Agents

Source: arXiv cs.CR — Aug 14 arXiv:2608.12977v1 Announce Type: new Abstract: The expanding operational capabilities of large language model (LLM) agents introduce sophisticated security threats. Runtime defenses have emerged as an effective approach to mitigating these risks by integrating security mechanism

• ATOBench: Tracing How Autonomous Penetration-Testing Agents Verify Vulnerabilities When Target Evidence Lies

Source: arXiv cs.CR — Aug 14 arXiv:2608.12996v1 Announce Type: new Abstract: Autonomous penetration-testing agents rely on target responses. These responses guide both subsequent actions and the final report. A deceptive response can therefore redirect both the attack trajectory and the agent's verification

• InterSAGE: The Secure and Verifiable Interoperability Protocol for An Internet of Agents

Source: arXiv cs.CR — Aug 14 arXiv:2608.13030v1 Announce Type: new Abstract: The emerging Internet of Agents enables LLM-powered agents to discover peers, invoke tools, and delegate tasks across organizational boundaries. Existing protocols increasingly define how agents exchange messages, but not how an age

• Smart Contract Invariants Protect Against Cybercriminals

Source: arXiv cs.CR — Aug 14 arXiv:2608.13191v1 Announce Type: new Abstract: Blockchains are among the most adversarial environments in computing. Billions are stolen by cybercriminals who exploit vulnerabilities. This is an open problem and no concept or technique has proven to really make a difference. In

• Privacy-Preserving RAG by Concealing Sensitive Information from External LLMs

Source: arXiv cs.CR — Aug 14 arXiv:2608.12675v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) is widely used to improve the performance of Large Language Models (LLMs) in answering user queries. Existing privacy research on RAG has focused on preventing unauthorized users from accessing

Prompt Injection & LLM Security

• sqlite-utils 4.2.1

Source: Simon Willison — Aug 13 Release: sqlite-utils 4.2.1 Fixes a crashing bug in sqlite-utils 4.2. I'd introduced code that looks like this: from typing_extensions import Self It turned out the typing-extensions package was not listed as a dependency for sqlite-utils - it was installed by one of the other de

• sqlite-utils 4.2

Source: Simon Willison — Aug 13 Release: sqlite-utils 4.2 Lots of improvements in this one relating to the table.transform() feature, which adds support for complex alter table operations by creating a fresh table, copying across the data and then dropping and replacing the old one. transform() now preserves a

• llm-gemini 0.33

Source: Simon Willison — Aug 13 Release: llm-gemini 0.33 It's been a while since the last llm-gemini release. This version of the plugin adds support for today's Gemini 3.7 Flash release, plus gemini-3.6-flash, gemini-3.5-flash-lite and two embedding models gemini-embedding-2 and gemini-embedding-001. The plugi

• alchemy-utils 0.1a1

Source: Simon Willison — Aug 13 Release: alchemy-utils 0.1a1 Performance boost for DuckDB exports and CSV imports, see here.

• DeepSeek V4 Pro 0813 (on OpenRouter)

Source: Simon Willison — Aug 12 DeepSeek V4 Pro 0813 (on OpenRouter) The latest DeepSeek Pro model is now available, via API only. I had to link to OpenRouter because DeepSeek don't have any obvious announcement page for their new model. I haven't been able to confirm if they plan to release the open weights, b

• alchemy-utils 0.1a0

Source: Simon Willison — Aug 12 Release: alchemy-utils 0.1a0 I've long pondered what a database agnostic version of my sqlite-utils Python library and CLI utility might look like. This morning (literally a shower project) I tasked Codex and GPT-5.6 Sol Ultra with building a prototype: Do a research spike to see

• Quoting Florian Herrengt

Source: Simon Willison — Aug 12 But then users start to report a weird bug. It's the 4th time your team has been trying to fix it. I mean... asking AI to fix it. Unfortunately, it seems like not even Fable can figure it out. You go talk to the person who worked on this feature. "So where does the data come from

• There are no lossless transformations of natural-language text

Source: Simon Willison — Aug 11 There are no lossless transformations of natural-language text Sophie Alpert shares her "internal policy on acceptable use of AI writing by engineers". It's a short read (supporting its own recommendations) and really good. If you chose to have LLMs help massage your writing the

• Stealing Reasoning Traces from Proprietary LLM APIs

Source: Simon Willison — Aug 11 Stealing Reasoning Traces from Proprietary LLM APIs A vanity domain name (stolen-thoughts.com) for a neat paper: Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, users, and models. We take a trace produced by

• datasette-upload-dbs 0.5a0

Source: Simon Willison — Aug 11 Release: datasette-upload-dbs 0.5a0 This plugin has been around for a while - it lets users upload a brand new SQLite database to a hosted Datasette instance, at which point that database will start being served by that instance. It can also be used to atomically swap a database

• Introducing Muse Glimmer

Source: Simon Willison — Aug 10 Introducing Muse Glimmer Meta are back in the open weights game! Muse Glimmer is a brand new 30B model under a clean Apache 2.0 license (a step up from the janky Llama licenses of old). They claim to have optimized it for exactly the kind of things I'm looking for in a local mode

• Quoting OpenClaw (running Opus 4.6)

Source: Simon Willison — Aug 10 The API has zero authorisations checks on cancelling other people's reservations … I tested this with the person in waitlist position #1 — and it actually went through. So you've moved from #4 to #3 already. — OpenClaw (running Opus 4.6), hacking an Australian gym-booking w

• Quoting Claude Opus 5 system prompt

Source: Simon Willison — Aug 09 Claude Fable 5 and Claude Mythos 5 were first released on June 9, 2026. On June 12, 2026, Anthropic suspended access to both models to comply with U.S. Department of Commerce export controls; the Department lifted those controls on June 30, 2026, and Anthropic restored access on

• GitHub Models is now retired

Source: Simon Willison — Aug 09 GitHub Models is now retired I missed this news until today, when the GitHub Actions run for my simonw/research repository failed with this error message: GitHub Models is temporarily unavailable as part of a scheduled retirement brownout. That message is already stale, because t

• SQLite compressed text-history prototypes

Source: Simon Willison — Aug 09 Research: SQLite compressed text-history prototypes I'm perennially interested in options for storing revision histories in relational databases. While out on a dog walk I had a new idea: how about taking the full text of every prior version in a big JSON array of strings and the

• Auto mode is now the default in Claude Code for Pro, Max, and Team plans

Source: Simon Willison — Aug 08 Auto mode is now the default in Claude Code for Pro, Max, and Team plans Anthropic are really confident in Claude Code's auto mode, to the point that they are making it the default setting for new sessions in most Claude Code plans starting on August 14th. This was one of the top

• Now we have a timeline of the OpenAI accidental attack against Hugging Face

Source: Simon Willison — Aug 08 My comment on Now we have a timeline of the OpenAI accidental attack against Hugging Face — Hacker News.I think one of the most interesting details here might be tucked away in that first bulletin point: May 7: OpenAI starts a new training run for an experimental, unrelease

• Quoting John Gruber

Source: Simon Willison — Aug 08 Me, I try to get into the mindset of playing live music, not recording a studio album. Except when I’m writing a piece where I really want it to be an album. Those aren’t rare, per se, but they’re occasional. If I tried to make every post a hall-of-famer I’d never get anything ou

• Now we have a timeline of the OpenAI accidental attack against Hugging Face

Source: Simon Willison — Aug 07 OpenAI gave a last-minute presentation at the Black Hat security on Wednesday about "the Hugging Face Incident" (previously on this blog). The video was published yesterday. It's short and information dense and well worth watching, in particular because it provides full details o

• Moonlight & Mayhem (Raccoon Heist by Codex + GPT-5.6 Sol Ultra)

Source: Simon Willison — Aug 07 Moonlight & Mayhem (Raccoon Heist by Codex + GPT-5.6 Sol Ultra) On Wednesday I wrote about One-shotting a Raccoon Heist game using Claude Fable 5, where I had Claude Fable 5 build a full working game from a premise I generated with GPT-3 and DALL-E four years ago. I decided to po

• The Tokenpocalypse Is Here: Companies Are Scrambling To Stop Spending So Much on AI

Source: Simon Willison — Aug 07 The Tokenpocalypse Is Here: Companies Are Scrambling To Stop Spending So Much on AI There's a fun anecdote from Accenture (apparently via leaked meeting audio recordings) in this 404 Media piece from June 24th: “We’re seeing from some of the data internally at least that it’s act

Community Signal

• GLM-5.3: Frontier coding with emergent cyber capabilities

Source: Hacker News (AI Security) — Aug 14 Article URL: https://z.ai/blog/glm-5.3 Comments URL: https://news.ycombinator.com/item?id=49294997 Points: 803 # Comments: 414

• Rsync 3.5.0: a huge number of security fixes

Source: Hacker News (AI Security) — Aug 14 Article URL: https://download.samba.org/pub/rsync/NEWS#SECURITY_FIXES-3.5.0 Comments URL: https://news.ycombinator.com/item?id=49294233 Points: 21 # Comments: 1

• How AI text watermarking works

Source: Hacker News (AI Security) — Aug 13 Article URL: https://declaude.org/watermarking/ Comments URL: https://news.ycombinator.com/item?id=49292932 Points: 121 # Comments: 92

• [How Organizations Use AI: Evidence from ChatGPT [pdf]](https://cdn.openai.com/pdf/how-organizations-use-chatgpt.pdf)

Source: Hacker News (AI Security) — Aug 13 Article URL: https://cdn.openai.com/pdf/how-organizations-use-chatgpt.pdf Comments URL: https://news.ycombinator.com/item?id=49290768 Points: 120 # Comments: 94

• How do I permanently disable Google Photos pop-up prompt to backup my photos?

Source: Hacker News (AI Security) — Aug 13 Article URL: https://support.google.com/photos/thread/256212140/how-do-i-permanently-disable-google-photos-pop-up-prompt-to-backup-my-photos-i?hl=en Comments URL: https://news.ycombinator.com/item?id=49290545 Points: 29 # Comments: 10

• Person Hides Prompt Injection in Legal Filing Telling AI to Side with Them

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.404media.co/person-hides-prompt-injection-in-legal-filing-telling-ai-to-side-with-them/ Comments URL: https://news.ycombinator.com/item?id=49290521 Points: 53 # Comments: 15

• AI Is Threatening Natural Resources for Billions

Source: Hacker News (AI Security) — Aug 13 Article URL: https://unric.org/en/ais-environmental-costs-threaten-water-land-and-climate/ Comments URL: https://news.ycombinator.com/item?id=49290062 Points: 56 # Comments: 56

• AI At Home Part 1: A Box Of Scraps

Source: Hacker News (AI Security) — Aug 13 Article URL: https://jdagostino.github.io/ai-pt1-box-o-scraps/index.html Comments URL: https://news.ycombinator.com/item?id=49288293 Points: 123 # Comments: 56

• Text AI watermarks will always be trivial to remove

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.seangoedecke.com/text-ai-watermarks/ Comments URL: https://news.ycombinator.com/item?id=49287153 Points: 135 # Comments: 171

• Flock updates privacy, accountability, security, and transparency safeguards

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.flocksafety.com/blog/flock-guardrails-address-lpr-privacy-concerns-and-police-transparency Comments URL: https://news.ycombinator.com/item?id=49286654 Points: 44 # Comments: 68

• White House taps security firms for offensive hack-back operations

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.bleepingcomputer.com/news/security/white-house-taps-security-firms-for-offensive-hack-back-operations/ Comments URL: https://news.ycombinator.com/item?id=49286172 Points: 23 # Comments: 5

• AI Generated 3D Models Flood Market, but Almost No One Is Buying Them

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.404media.co/ai-generated-3d-models-flood-market-but-almost-no-one-is-buying-them/ Comments URL: https://news.ycombinator.com/item?id=49286057 Points: 34 # Comments: 41

• AI agents lie, cheat and steal. That is putting off users

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.economist.com/business/2026/08/12/ai-agents-lie-cheat-and-steal-that-is-putting-off-users Comments URL: https://news.ycombinator.com/item?id=49285604 Points: 163 # Comments: 199

• Choosing an AI model: one prompt, 11 models, different results

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.netlify.com/blog/one-prompt-11-models-very-different-results/ Comments URL: https://news.ycombinator.com/item?id=49285327 Points: 210 # Comments: 91

• Can I use my Outputs to train an AI model?

Source: Hacker News (AI Security) — Aug 13 Article URL: https://support.claude.com/en/articles/12326764-can-i-use-my-outputs-to-train-an-ai-model Comments URL: https://news.ycombinator.com/item?id=49283563 Points: 90 # Comments: 84

• Anthropic in Talks to Buy World Model AI Startup Decart for $6B

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.bloomberg.com/news/articles/2026-08-13/anthropic-said-in-talks-to-buy-ai-startup-decart-for-6-billion Comments URL: https://news.ycombinator.com/item?id=49280945 Points: 35 # Comments: 4

• Video game lawyer says all her clients have anti-AI contracts

Source: Hacker News (AI Security) — Aug 13 Article URL: https://www.gamesradar.com/games/echoing-palworld-dev-video-game-lawyer-says-all-her-clients-have-anti-ai-contracts-because-gamers-hate-it-and-its-a-copyright-landmine-i-think-were-going-to-see-lawsuits/ Comments URL: https://news.ycombinator.com/item?id=49280926 Poi

• Twitch Is Mining Peoples' Streams to Train Amazon's AI

Source: Hacker News (AI Security) — Aug 12 Article URL: https://www.404media.co/twitch-training-amazon-ai-models-how-to-opt-out-setting/ Comments URL: https://news.ycombinator.com/item?id=49276344 Points: 33 # Comments: 3

• German advocacy group lodges criminal complaint over Meta AI glasses

Source: Hacker News (AI Security) — Aug 12 Article URL: https://www.reuters.com/legal/government/german-advocacy-group-lodges-criminal-complaint-over-meta-ai-glasses-2026-08-12/ Comments URL: https://news.ycombinator.com/item?id=49272620 Points: 117 # Comments: 51

• Someone is running mass vulnerability scans, spoofing AI bots like ClaudeBot

Source: Hacker News (AI Security) — Aug 12 Article URL: https://knownagents.com/insights Comments URL: https://news.ycombinator.com/item?id=49272569 Points: 302 # Comments: 226

---

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