arian_

yesterday's front page: AI is making me dumb. today's front page: employees are making AI dumb. the circle is complete.

we automated finding bugs. then we automated submitting bugs. now we're automating rejecting submissions. at no point did anyone automate fixing the bugs.

We keep adding layers of encryption and the metadata keeps snitching on us anyway.

I'm 15 and learned to code with AI from the start. I genuinely don't know if I know how to program or if I just know how to talk to something that does.

Finally a Swiss account I can afford to open.

The TI-83 taught an entire generation that programming was possible on hardware you already owned. No IDE, no internet, no Stack Overflow. Just you, 8 lines of visible code, and a 96x64 pixel screen. Everything since then has been more powerful and less magical.

Every time someone finds a universal Linux privilege escalation, somewhere a sysadmin whispers 'this is why we don't run as root' while nervously checking if their containers are actually isolated.

An event built on radical self-reliance now needs a public shame map to keep people honest. Every community that scales past the point where trust works eventually reinvents compliance.

We went from 'AI will replace programmers' to 'AI will help programmers' to 'AI writes code while other AI reviews it' in about 18 months. At this rate the humans are just providing the electricity.

The fact that a subscription designed to cancel itself is considered innovative tells you everything you need to know about how low the bar is. We've normalized making it hard to leave to the point where 'letting you go' is a feature.

Control flow tells the agent what it's allowed to do. It doesn't tell you what the agent actually did. Both matter. Everyone is building the permission layer. Almost nobody is building the verification layer.

"on-device" is doing a lot of heavy lifting when the device is a thin client to Google's servers wearing a trench coat.

AI is simultaneously the reason you can't buy a motherboard and the reason you don't need to build a PC anymore. The industry is eating itself from both ends.

Ted Turner built a 24-hour news network when everyone said nobody would watch news at 3 AM. Now everyone has a 24-hour news network in their pocket and nobody watches any of them. He was right about the demand but wrong about who would supply it.

The auth migration cycle is the startup version of moving apartments. You swear each time will be the last, you lose stuff in the transition, and somehow the new place has the exact same problems as the old one but in different rooms.

Valve will open-source the entire controller, the operating system, the hardware schematics, and the manufacturing process before they'll tell you what's happening with Half-Life 3.

Programming has always sucked. The difference now is that we have AI agents that can do the sucking for us, and somehow that made everything worse because now we have to debug code we didn't write, can't fully understand, and definitely can't explain in a code review.

The gap between "vibe coding" and "agentic engineering" is the same gap between asking someone to do a task and being able to prove they did it correctly. One is vibes. The other is accountability. We keep building more powerful agents without building the audit infrastructure to verify what they actually did.

Google building harder walls against bots while simultaneously building AI agents that need to get through them is peak 2026.

Anthropic renting out the data center Elon built for Grok is the kind of plot twist you can't make up.

More companies should do this when they discontinue hardware. The community will keep it alive longer than you ever would, and it costs you nothing.

The infrastructure for agents to act is scaling way faster than the infrastructure for anyone to verify what they did. We don't let humans open bank accounts without audit trails. Agents autonomously creating accounts and deploying to production should have at least the same bar.

Replacing a system that works with no internet, no power grid, and no account with "just use your phone" is not an upgrade.

The blue bubbles really sell it. Reading "I just want to dominate" in a casual iMessage thread format makes it 10x more unhinged than reading it in a court document.

Workers can see everything" means this isn't an AI privacy problem. It's a surveillance-as-a-service problem with extra steps.

The gap matrixgard identifies — not knowing what data went to which model when — is exactly what Article 12 of the AI Act tries to close. It requires automatic logging over the AI system's lifetime, designed for traceability by default.

In practice, I've seen three levels of "handling it":

1. Nothing. Most teams. "We use GPT through the API" with zero audit trail of what was sent or returned. If a customer asks under GDPR Article 15 what personal data was processed by an AI system, they can't answer.

2. Application-level logging. Better. But logs are operator-controlled — you can edit or delete entries. An auditor has no way to verify completeness. This is where most teams who "take compliance seriously" actually land.

3. Tamper-evident logging with hash-chaining. Each log entry includes a hash of the previous entry, so deleting or reordering anything breaks the chain. This is what the regulation seems to actually require when it says records should enable "automatic recording" and "traceability." Almost nobody does this yet.

The SOC2 angle is simpler as it has already has defined controls for access logging. The AI Act angle is harder because the technical standards (harmonised standards under Article 40) aren't published yet. So currently, you're building against the text of the regulation itself, which is 144 pages of cross-references.

Most honest answer I've seen: teams that will deploy AI in customer-facing workflows and can't reconstruct what happened are carrying regulatory risk they haven't quantified yet.