RC RANDOM CHAOS

Articles

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

A claim, not a control
GrapheneOSmobile security

A claim, not a control

GrapheneOS rewrote its Messages app under a privacy and security claim, but the enforcing controls are not named or independently verifiable.

8 min read
Forty lines freeze your Mac
macOS securitydenial of service

Forty lines freeze your Mac

How the Deathray technique lets an untrusted website freeze an entire Mac by overloading WindowServer, why the browser sandbox can't stop it, and what to do.

7 min read
Google Ads became a malware distribution channel
malvertisinggoogle-ads

Google Ads became a malware distribution channel

Malvertising turns Google's top search ad into a malware delivery channel. How the attack works, why review misses it, and the steps that stop it.

7 min read
HuggingFace serves a disclosure channel, not a control
security.txtvulnerability disclosure

HuggingFace serves a disclosure channel, not a control

HuggingFace's security.txt defines a disclosure channel, not a security control. Publishing it expanded the commitment surface, not the platform's defenses.

8 min read
Keep the two claims apart
AI ethicsdata privacy

Keep the two claims apart

Consumer AI trains on your chats by default. How to tell the real consent problem from unprovable 'secret breakthrough' claims - and what you can control.

7 min read
Microsoft's new default is Rust
Rustmemory safety

Microsoft's new default is Rust

Microsoft made Rust a tier-1 language because memory-safety bugs drove ~70% of its yearly security patches. Here is what the shift means.

7 min read
The model kept no receipts
AI safetyprovenance

The model kept no receipts

The "OpenAI stole my proof" fight is really a provenance gap in AI, and that gap is a safety problem, not just a credit dispute.

7 min read
Turning off training does not protect your data
LLM engineeringdata governance

Turning off training does not protect your data

OpenAI keeps re-enabling 'allow training.' Why a UI toggle is not a data control, and the gateway-and-contract architecture that makes training exposure impossible by design.

10 min read
Desert Ant Labs puts inference inside the loop
local LLMsAI pipeline architecture

Desert Ant Labs puts inference inside the loop

Fast local models turn inference into a near-zero-cost function call - here is the tiered pipeline pattern, a real support-desk example, and where it breaks.

9 min read
SWE-2 is the wrong model for almost everything
SWE-2agentic pipelines

SWE-2 is the wrong model for almost everything

SWE-2 is a code-execution node, not a generalist replacement. How to eval it against your own repos and where it actually fits in production pipelines.

10 min read
The weaker model matters more than the smarter one
LLM engineeringDeepSeek v4.1 Flash

The weaker model matters more than the smarter one

DeepSeek v4.1 Flash cuts token cost, not the need for validation - use it in cascades, verification loops, and long-context pipelines that hold up in production.

10 min read
Your defenses are now executing for the attacker.
forgejorce

Your defenses are now executing for the attacker.

Forgejo 16.0.3 and earlier carry a critical RCE. The vulnerable version is the exposure, and running a build above the range is the only confirmed mitigation.

7 min read