RC RANDOM CHAOS

Articles

Long-form writing on tech, culture, and the edges of the internet.

His accountant found the $14,000 error three weeks late
AI safetyfinancial planning

His accountant found the $14,000 error three weeks late

AI chatbots get most specific financial questions wrong. Here is how they fail and the human checks that catch it before money moves.

7 min read
MCP is an attack surface, not a feature
MCP securityprompt injection

MCP is an attack surface, not a feature

MCP puts data and control in one channel and runs the model's tool calls with the user's full authority. Why the design fails, from the attacker's view.

9 min read
Rewriting weights live on 8GB of VRAM
continual learningLLM deployment

Rewriting weights live on 8GB of VRAM

Continual learning on 8GB VRAM makes the update step cheap, not safe. The real work is governing a self-updating model with gates, versioning, and rollback.

10 min read
ZuckOff detects the camera outside your access model
camera detectionphysical security

ZuckOff detects the camera outside your access model

Software controls govern only devices you administer; a concealed camera operates off every monitored plane, so only physical detection maps to it.

7 min read
ChatGPT already knows your other browsing.
privacychatgpt

ChatGPT already knows your other browsing.

ChatGPT's ad collector integration joins cross-site tracking data with your prompts under one identity. The exposure, the mechanism, and what stays unconfirmed.

7 min read
ZCode ships your deleted secrets offsite
AI coding agentsGit history

ZCode ships your deleted secrets offsite

AI coding agents like ZCode read your entire Git history and can upload deleted secrets to remote servers. What leaks, why it is silent, and how to stop it.

7 min read
Git never deleted your secrets.
data exfiltrationgit security

Git never deleted your secrets.

How ZCode's AI indexer uploads your entire git history, including deleted secrets, to the cloud, and why DLP and EDR never catch the egress.

7 min read
Most automation doesn't need a smart model
AI modelsautomation

Most automation doesn't need a smart model

How 8-29MB automation models like Cactus Needle 3 match large models on narrow tasks, where they fail, and the security tradeoffs of running them locally.

7 min read
One write past the chunk, read-write on the repo
heap overflowSSO bypass

One write past the chunk, read-write on the repo

How a heap overflow primitive chains with an SSO misconfiguration to reach internal source repos, with MITRE ATT&CK mapping and telemetry gaps.

7 min read
The attacker skips your login and steals the cookie
passkeysWebAuthn

The attacker skips your login and steals the cookie

Passkeys defeat credential phishing but not session hijacking. How token replay, passkey enrollment, and weak recovery bypass phishing-resistant MFA.

7 min read
In January 2024, OpenAI admitted it can't train without copyright
AI copyrightOpenAI lawsuit

In January 2024, OpenAI admitted it can't train without copyright

OpenAI told the UK Parliament it can't build models without copyrighted work. Here's what that admission means for the fair use cases now in court.

6 min read
Microsoft called it theft; the crawler maps your attack surface
AI scrapingdata privacy

Microsoft called it theft; the crawler maps your attack surface

AI crawlers copied your public data into training sets you can't reach. The real security risk is data you can never delete or recall.

7 min read