AI Micro-Notes 2026: Thoughts Too Short to Trash

A curated set of short AI hot takes I’ve been posting on Threads since 2026, now organized by theme rather than by month. Some are too short to turn into a full article, but the opinions or roasts feel too good to throw away. The more scattered, time-sensitive notes moved to the 2026 chronological archive; for 2025 takes, see the 2025 archive.

Model Temperament, Pricing & Quota

Knowledge Cutoff

Watch out for knowledge cutoff when vibe coding. If you specify Gemini 3 Pro, the model might think it doesn’t exist yet and quietly swap in an older version.

Local vs Cloud Models

Cloud models? 8GB is enough to run them. Local models depend on your specs — but why go local? If it’s for privacy, that’s fair. If it’s to save money, not really — because making a local model perform on par with commercial ones means shelling out for hardware first. My advice: don’t overthink local early on. Get comfortable with closed-source models first.

Claude Import Mode for GPT Memories

Claude has an import mode, right? You can extract GPT’s knowledge and memories about you, then import them into Claude.

Recitation Error Solution

Gemini is the strictest on recitation errors, Qwen is the most lenient. So Qwen via US-based providers on OpenRouter is the best choice — data doesn’t flow to China since it’s deployed in US data centers, and the model only restricts politically sensitive content, not copyright.

The Ones Who Roast Claude Hardest Use It Longest

The people who roast Claude the hardest are the ones who’ve used it the longest.

Sonnet Is Becoming the New Haiku

Anthropic’s latest article listed Fable, Opus, and Sonnet, but never mentioned Haiku. Sonnet appears to be taking Haiku’s old place. That is a quiet price increase, and Haiku may well stop at 4.5.

Dev & Security Pitfalls

rm -rf Warning

There are already plenty of horror stories about rm -rf online. Never run that with dangerously skip permission enabled.

Plaintext API Keys Sitting in Codex Snapshots

Scanned ~/.codex/shell_snapshots/ and found one snapshot exporting 16+ API keys in plaintext, with 0644 world-readable file permissions.

Two Traps in Vercel Throwaway Public Pages

A fresh project’s deploy URL ships with Deployment Protection (Vercel Authentication) on by default — outside visitors get a 401. The other trap: the bare <proj>.vercel.app alias may be someone else’s empty shell; your auto-updating prod alias is <proj>-<team>.vercel.app. curl and check the content before sending anyone the link.

A File Named “Transcript” Isn’t Necessarily a Transcript

The file was named “transcript,” but it turned out to be an already-edited summary — the raw recording lived elsewhere. Bonus trap: SenseVoice STT has no speaker diarization, so in multi-speaker sessions the opening introductions get attributed to the wrong people.

Three Verification Misalignments

Verifying against your own spec is worse than verifying against the actual computation — if your spec has an error, you’ve made your own mistake the baseline. Deep-research false negatives: 16/19 in the “refuted” bucket were 0-0 or 1-0 votes (agents timed out without verifying, not actually refuted). Dry-runs validate data integrity but can’t catch an exercise that simply doesn’t work — you need to simulate a real student’s spoken prompts with step-by-step chaining to surface that.

iQOS Bluetooth Reverse Engineering

I’m genuinely impressed by Claude Code. I saw someone on Reddit share how Claude Code helped them defeat ransomware and recover data. On a whim, I plugged my iQOS into the computer and asked if it could read the data. It actually went online to research, found an open-source reverse engineering project, read through the logic, then wrote its own script to pull data from the iQOS via Bluetooth.

Workflow & Method

CLAUDE.md Prohibition Paradox

I wrote in CLAUDE.md: “When corrected by the user, never say ‘you’re right.’” Claude immediately ignored it. After I called it out for violating the rule, its next response was: “You’re right, I shouldn’t have violated the language prohibition.” CLAUDE.md is advisory, not enforced — if you really need to block something, use Hooks.

Sonnet 4.6 Plus MCP for Suspicious Emails

Ordinary folks don’t need to reach for Mythos — Sonnet 4.6 wired up to an MCP can help analyze suspicious malicious emails too. “The email was crudely made, the template placeholders weren’t even swapped out” — I wouldn’t have noticed that if it hadn’t pointed it out. Just remember to tell the model not to click unfamiliar links, unless you have a proper sandbox.

How Someone With GAD Reads a Risk Report

Let me show you how I, with generalized anxiety disorder, read a risk report: One, an 8.6% risk is way too high, unacceptable. Two, if it happens, 70% of cases will match or exceed what I imagined? Unacceptable. Three, we never just worry about 30 days out — 30 years sounds more like it.

First-Principles Skill Cuts to the Essence

Whenever everything gets tangled up, I reach for my Musk first-principles skill. It usually cuts straight to the essence and strips away the unnecessary branches.

Don’t Mistake a Third-Party Feature for Official Claude

A “new Claude feature” going viral was actually built by a company called fastlane — not Anthropic, so it doesn’t count as official at all. If I didn’t watch Anthropic like a hawk every day, roasting them, knowing their brand style and (fake) moral fastidiousness inside out, I’d have been fooled too.

Automation Zealots, Refueling the Fire Daily

There’s a team that champions automation to the extreme — the bald guy’s CC team, and their company. The result: they blow up every single day. I figured the humans would step out to wrap things up; instead their people step out to pour fuel on the fire.

Route Social Sites by Tool Capability

Direct WebFetch from Claude or Codex often gets blocked on X and Threads, so they need a browser detour. Gemini can read them directly. Reddit is especially hostile to Claude while Codex and Gemini usually work. Model choice is also about access, not just intelligence.

Polite Does Not Mean Inventing an Excuse for Them

“I know things have probably been busy on your end” sounds considerate, but it invents a reason for someone else’s silence. AI-written client emails often hide this kind of condescension. I still review every draft after running my polite skill.

AI Industry & Business

Anthropic in China

Search for Claude/Anthropic on Xiaohongshu (China’s Instagram-like social platform) and you’ll find it’s one of the few companies that once explicitly “insulted China” but came out unscathed. Now all you see is “it’s so good / how to use a VPN to access it.” Strength is the ultimate argument.

Claude Moves Wall Street

Claude is the one AI that can single-handedly shake the stock prices of major software companies on Wall Street. Enough said.

Karpathy — The Ultimate Free Agent

Andrej Karpathy announced he’s joining Anthropic. OpenAI to Tesla to OpenAI (return) to departure to Anthropic. Probably the most prolific team-switcher in AI. Then again, top talent gravitating toward where they see the most potential is itself a market signal.

Learn Slow Enough and You Learn Nothing

In the AI era, if you learn slowly enough, you can skip learning anything at all. By the time you finish, that thing has already been automated. Flip side: people who learn fast enough keep finding they can do more and more.

Benchmark Chasing vs Alignment

Most models are focused on chasing benchmark scores. But there are aspects I think deserve more attention: human intent alignment — understanding what I want from minimal input; constraint compliance — when I say don’t do something, don’t do it; and actually useful context length.

The AI-Course Cash Grab, Overheard at a Convenience Store

Sheltering from the rain at a 7-Eleven seating area in Da’an, Taipei, waiting for class, I overheard some aunties talking about AI: “I’m paying NT$18,000 a month for my son to learn AI, four installments now, it’s that teacher’s course he recommended last time.” Infuriating — I really want to know which teacher rakes it in this well. The market is now buzzing with “Law of Attraction AI,” “Benefactor AI,” “share your gratitude and closed deals with AI,” “AI raises your personal energy,” “networking coach radiating beauty” — I can’t take it anymore. The highest-value overheard conversations really do happen in convenience stores and fast-food seating areas.

A Vote for Gemini: Let the Facts Talk

In an online argument, who do I back? Gemini — state the facts plainly, leave the rest to the onlookers, let the facts do the talking. That’s enough (you should never expect to convince the other side). Onlookers don’t need a pile of agitation and emotional payoff.

L3 BD Outreach Open Rate

First email using “let’s exchange notes” framing (no pitch, no needs ask, just “want to grab some time, exchange notes on AI in practice”) got 67% open rate. Same list with a “let me sell you” framing (course pitch, L3 consulting service) got 17%. Four times the gap. Don’t sell anything in the first email — that’s actually true for the early stage of a BD funnel.

The Two Sides of Anthropic’s Need to Teach

During a PR crisis, Anthropic’s “let us educate the user” posture is unbearable. The same instinct also produces unusually careful technical writing. Unknown Unknowns and the recent piece on effort are both excellent.

Life & Miscellany

Coming Home, I Realized Taiwan Got Rich

Back in Taiwan this trip, it really feels like the place got rich — in hospitals, restaurants, on the metro, the overheard chatter is all about buying stocks and investing; office workers in the elevator talk about trips to Japan and Europe. A set meal at my usual spot went from NT$160 to NT$220 in two years; an Uber that used to be NT$160-200 is now NT$250-300. Living in Thailand, I can barely keep up.

A Sense of Boundaries

Learn one more term while you’re at it: a sense of boundaries. Many say it’s a Mainland Chinese coinage, but I find it irreplaceable. The relative you haven’t seen since last New Year prying into your private life — how much you earn, are you married yet — that’s the absence of boundaries.

A Wuxia-Style Water Dispenser

Deep winter, snow falling, the swordsman arrives at the inn, parched. Innkeeper: please have some warm water. “Have you boiling water?” Innkeeper: first press unlock, then press hot.

Vietnamese Coffee Shop Internet

When I first arrived in Vietnam, I avoided those open-front street coffee shops with low tables and camping chairs, assuming the internet would be unreliable. Turns out these places have the best connections — because Vietnamese teenagers camp there all day playing mobile games on a single coffee order. If the latency were high, nobody would show up.

First Principles: If Holding a Grudge Doesn’t Help, Delete It

First principles, right? Does holding a grudge make things better? No. Then delete it.

The AI-Detects-the-Fallen Utopia Is Still Far Off

Someone wished AI could detect a collapsed person in a riverside park and auto-call for help. The problem: covering an entire riverside with detection is basically Skynet — the privacy backlash would explode, and it’s economically unviable. Even Tesla FSD is road-legal in only a handful of places worldwide. The pragmatic option today is to wear the detector yourself (Apple Watch fall detection auto-dials emergency). It’s not that Taiwan lacks it — nowhere in the world has it yet.