RC RANDOM CHAOS

Articles

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

The benchmark score is the number to trust least
AI safetyClaude Opus 5.5

The benchmark score is the number to trust least

How to weigh Claude Opus 5.5's intelligence, latency, and token cost, and where its real AI safety and cybersecurity risks concentrate.

7 min read
the phone rings in a voice you trust
deepfake voice cloningsocial engineering

the phone rings in a voice you trust

Unreal Agent clones trusted voices for social engineering, and any identity check that ends at voice recognition is already defeated.

6 min read
Your telemetry toggle was a file access control
least privilegeaccess control

Your telemetry toggle was a file access control

Claude Code read AGENTS.md only when telemetry was on. Binding access to a diagnostic flag breaks least privilege. The fix closed the instance, not the class.

7 min read
An unsolved cipher was never a secure one
AI cryptanalysislegacy encryption

An unsolved cipher was never a secure one

GPT-6 Astra solved an Enigma message unsolved since 2005. The failure was effort-based security, not the cipher, and that pattern reaches legacy encryption.

6 min read
Every Prompt Is A Retrieval Request
AI model securitydata exfiltration

Every Prompt Is A Retrieval Request

Meta's Muse returned 6.8GB when asked for its filesystem. The response channel serviced a request for internal data with no enforcement at the boundary.

6 min read
Grok 4.7 collapses recon into a scripted session
threat-intelligencedetection-engineering

Grok 4.7 collapses recon into a scripted session

Grok 4.7's real-time retrieval and agentic tool-use compress red team reconnaissance and phishing, degrading content-based detection while behavior-based rules hold.

6 min read
Heretic strips refusals from open-weight models
AI safetyabliteration

Heretic strips refusals from open-weight models

Heretic automates stripping refusals from open-weight LLMs. Why model-level guardrails were never a security control, and what defenders should do instead.

6 min read
Mini-AGI trains a model you can't sign
AI safetycontinual learning

Mini-AGI trains a model you can't sign

Mini-AGI trains a continual-learning model on 8GB VRAM. Why self-updating weights break signing, auditing, and red-team results, and how to defend.

7 min read
One ../ and the server hands over the wrong file
path traversalconditional rce

One ../ and the server hands over the wrong file

Unauthenticated WordPress path traversal lets an anonymous request select which file the server reads; execution follows where a code-executing sink exists.

7 min read
Your watermark is a spymark
spymarkwatermarking

Your watermark is a spymark

Watermark detection recovers a signal, not an author, so any party with write access can embed a hidden tracking mark defender tooling reads as clean.

8 min read
A privacy toggle is a sign, not a lock
ai privacygrok 4.7

A privacy toggle is a sign, not a lock

A security analysis of Grok 4.7's privacy versus other AI assistants, judged by enforcement point rather than feature count.

8 min read
Cloudflare ships Python Workers to general availability
python workersautomation pipelines

Cloudflare ships Python Workers to general availability

How to wire Cloudflare Python Workers into automation pipelines as validated, Queue-connected stages for reliable, scalable AI systems.

10 min read