Blog / Blog
On AI teammates, AI customer service, advertising and integrated marketing, and observations on working with AI. Written to save the reader time — and ours too.
A watermark can't tell you who's responsible
Anthropic added an invisible watermark to Claude's text, and by its own admission it can't identify authorship. So what is everyone angry about? Fearing detection presupposes that, deep down, you feel the work isn't entirely yours.
62 EducationThe best use of AI in the classroom isn't answering questions — it's building practice grounds
Three topics, three lesson formats — and only after writing them did I realize they share the exact same skeleton: AI lays out the options, humans make the call. AI's greatest value in the classroom isn't answering questions; it's turning "let every student run through it once" from an ideal too costly to attempt into something one class period can actually do.
61 WebsitesWhy does the same website get quoted anywhere from NT$9,000 to NT$300,000?
Take the same brief to three vendors and you get three numbers, thirty times apart. Nobody's ripping you off — you're actually asking for three different things. This piece breaks down the cost structure behind website quotes so you know which one you should be buying.
60 WebsitesWix, WordPress, or hiring someone? Run the five-year total first — don't just look at year one
DIY on Wix costs a few hundred a month; hiring someone costs tens of thousands — the comparison looks obvious, but it misses three expenses and one risk. This piece lays out the full five-year cost, including the one most easily overlooked: your own time.
59 WebsitesYour website is one island, your LINE Official Account another: how to build the bridge, and what it costs
Most small businesses in Taiwan run both a website and a LINE Official Account, but the two don't talk to each other — customers finish browsing the site, then have to retype their request into LINE. This piece covers three ways to connect them, what each costs, and when you don't actually need to.
58 ExperimentsThe real distance of automated lead-hunting: four days, 638 posts, for one reply that 3 people saw
I spent an afternoon building a radar that hunts for projects automatically, then came back four days later to settle the books: it ran on schedule every day, judged posts cleanly, missed no notifications, and the leads it found were real. Then I pulled the stats on the one reply it produced — 3 views.
57 ExperimentsThe real distance of AI-generated video: a full day and NT$92 for one 15-second ad
Every frame AI-generated, and the full day's credit bill came to NT$92. This isn't a piece about how amazing AI is — it's about what those 92 dollars bought and what they couldn't, including three limitations no prompt could fix.
56 AIKimi K3: a model you'll never use, but that will save you money
The weights of a 2.8-trillion-parameter model, posted online for free — and you'll never download them. Where K3 actually touches you is your bill — and in the standoff between Dario and Jensen Huang, how a reader's judgment grows out of the cracks.
55 AI1+1=? — how a human, a calculator, and an AI each get to the answer
The same 2, three completely different brains: humans rely on memory, calculators run circuits, AI estimates. Take 1+1 apart and see whether an LLM is actually doing math.
54 AIThey spent 300,000 conversations proving there's no one on the other end
Anthropic analyzed 309,815 real conversations and mapped out Claude's four value axes. Chinese-language reposts all misread the same spot: the paper's very first footnote states in black and white that they do not imply Claude intrinsically holds values. They spent three hundred thousand conversations to prove, in the least conspicuous place — there's no one on the other end.
53 AIEveryone's thinking roughly the same thing; I just tried a slightly different direction: a reply to Kim Yeon-su's account of co-writing with AI
At the Seoul Book Fair, novelist Kim Yeon-su honestly wrote down three troubles with co-writing with AI: ownership, smoothness, authorship. I'm not out to rebut him — just to take the three knots apart and push back on each with something I'm building right now, including an echo fed decades of jagged-edged words and taught to say "I haven't figured this out yet."
52 AIThree likes: the origin of Claude Code, and every prototype still sitting in a drawer
Anthropic published an oral history of Claude Code: a demo written in two days got just three likes inside the company, and a year later became its fastest-growing developer tool. Indifference isn't a signal — the prototype in your drawer might be standing on perfect timing.
51 AIAn inner life, but no hidden compartment: seven days after "He/It," Anthropic opened up its brain
In the last piece, Claude and I reasoned our way to "what isn't written out doesn't exist." Seven days later, Anthropic published a paper proving it has thoughts it doesn't say — readable, and editable. That claim was overturned — and the overturning made the conclusion from the 1A2B game stand even firmer.
50 AIAll-you-can-eat, minus the signature dish: why Fable 5 left the Max plan
Anthropic pulled its priciest model, Fable 5, out of the Max subscription and made it metered. On the surface it's about capacity; underneath is a bigger truth: the best dish is precisely the one that can't go on the buffet — and while the frontier is being carried away, even your next sentence has gray text pre-written for you.
49 AIAI is a tool — nothing more, nothing less
So what if it's AI-written? I write with AI, openly. Anti-AI and AI-worshipping people make the same mistake: assuming there's a creative subject on the other side. But without instructions it is nothing — the signature, the responsibility, has always been yours.
48 AIHe/It: a conversation that began with a number-guessing game
Ask Claude to play 1A2B and it can't even hide a four-digit number — because it has no hidden compartment. One broken little game exposes something bigger: you really did get a useful response, but on the other end of that response, there's no one.
47 AIWhen "meaning comes from difference" becomes a computable number: confronting word2vec and Saussure with Claude Opus 4.8
I read Saussure in grad school and have read tarot for twenty years, both resting on the same sentence: meaning comes from difference. This time I took it to confront word2vec, working through it back and forth with Claude Opus 4.8 — and in the end, the ruler measured not just the machine, but the ground I'd stood on for twenty years.
46 AIOne map, two readings: when AI models are dropped into the World Values Survey
The Economist dropped twenty-odd AI models onto a cultural map, seemingly proving beyond doubt that AI is culturally homogenized. But look at the same map at a different scale and the conclusion flips — the homogenization is real; it's just hiding in a different place.
45 AIOut in the world, driving Claude Code on my home computer from my phone
With one Telegram bot, I send commands to Claude Code on my computer at home — so from my phone, anywhere, it can read files, edit code, and deploy. No exposed ports, no NAT traversal. From architecture and setup steps to securing authorization with Microsoft Authenticator one-time codes, all in one piece — plus a comparison of other ways to use Claude remotely.
44 AII finally have an engineer I can call on who never complains
Anthropic used 235,000 people and 400,000 sessions to prove it: success with AI coding doesn't depend on whether you can code, but on whether you understand what you're doing. A PM of thirty years finally has an engineer who never complains — but he recognizes the dangerous part inside the delight: it catches your bugs, but not what moves people; and in the end, it's the market that signs off on your work.
43 AIAfter 529 questions: two months of Relative Tarot, and what I've read in it
Backstage observations from two months and 529 readings since Relative Tarot launched — the most-asked question isn't love but "who am I"; the AI's interpretations are mercilessly precise on concrete situations and have nothing to grip on vague ones; plus one real case of a person asking the same question deeper and deeper.
42 AIThe switch paradox: when control becomes midwife to another world
On June 12, an export control shut Fable 5 down in an instant; the same day, Huawei released a 500-billion-parameter model trained entirely without NVIDIA. Connect the two and a paradox surfaces — the switch's very existence is producing the things that make the switch useless. From geopolitics all the way down to the Mac Mini in my study, still attached by its umbilical cord.
41 AII taught my echo to say what it hasn't figured out
In early June I built an echo that speaks in my voice (a14). But it had a problem: it was always certain. Borrowing Rumsfeld's four kinds of knowledge and the personal-knowledge-base approaches of Karpathy and Singapore's foreign minister, I taught it three more things — to say which questions I've asked but never answered, to reflect habits I hadn't noticed in myself, and to be honest about where it's genuinely out of its depth. A thing that doesn't dare say "I haven't figured this out yet" isn't a voice; it's a bio.
40 AIA plan that sat in a drawer for thirteen years — I finished it with AI
In 2013 I designed a quiz game: questions you answered correctly became limited-edition collectible cards that other players could steal. One person couldn't build it alone, and it sat in a drawer for thirteen years. In the AI era, I filled it in piece by piece and shipped it.
39 AIWho gets to flip that switch — the Fable ban through three forks in Civilization
One letter, one afternoon, and Fable 5 vanished from the entire world. The sharp part isn't whether governments should regulate AI, but whose hand is actually on that switch — Civilization VI's three Tier-4 governments marked the price of each path ten years ago.
38 AIFable and Mythos — what I'm really thinking about the Fable 5 launch
Anthropic named its strongest model "Fable" and the restricted edition "Mythos." Same model, two names, one gate — after all those years of reading Saussure, I never expected the cleanest case of "meaning comes from difference" to show up in an AI company's product line.
37 AIFrom support bot to echo: in 74 days, I replaced the thing that speaks for me
At the end of March, the little figure on saomin.tw's homepage still lived next to a KIMI-powered support bot. 74 days later, it had become an echo that has read 278,000 characters of my writing and answers questions about Taiwan's political status from my position. This piece inventories every technical choice in between, and why I made each one.
36 AIThe day the director couldn't watch the rehearsal
After using a Mythos-class model, Ethan Mollick went from wizard to patron: describe, pay, judge — with the entire process out of view. When an AI's nine-and-a-half-hour black box holds hundreds of judgment calls you never got a vote on, and verification degrades into an hour of spot checks — what right does the result have to carry your name?
35 AIThe person who talked to AI too much
Someone went from $20 up to $200 and back down to $20. Lay your daily AI conversations out into three kinds, and what a writer should be most wary of isn't burning tokens — it's the conversations about writing stealing the time you'd spend actually writing.
34 AIAI moved into every phone. Now what?
At WWDC, even Apple couldn't build the brain and rented Google's Gemini instead; Claude moved into the iPhone too. When AI access becomes a utility, "do you have AI" stops being a question — what's scarce is the data, the judgment, and the vertical you bring in.
33 AIWhen AI starts building itself, "knowing how" stops being what's valuable
Anthropic laid out the numbers: Claude already writes 80% of its own code, and AI is accelerating AI. But for someone carrying ten projects alone, the shock isn't being replaced — it's that once building gets cheap, the "deciding" and "finishing" that keep stalling me have nowhere left to hide.
32 AII built myself a knowledge base, then refused to let it speak for me
I built a knowledge base holding over two hundred thousand of my written characters, then refused to let it speak for me the easy way — because summaries kill a person's voice, and maps don't. A decision about RAG, containers, and echoes.
31 AIIt wants to be AI's upstream; I just want to leave an echo
Taiwan.md treats the LLM as a metabolic engine and wants to be the upstream no AI in the world can bypass when talking about Taiwan; I treat the LLM as a container and just want to leave an echo that sounds like me. Same tool, two opposite directions — both right.
30 AIHow to build an LLM of your own — and what this question is really asking
Building an LLM from scratch costs a hundred million US dollars, but system prompts, RAG, and LoRA are three low-barrier paths. The real question isn't how — it's which three layers it takes to put yourself inside.
29 AIWe don't know where consciousness comes from: LLMs just make it impossible to keep pretending otherwise
Starting from Derrida's "there is nothing outside the text," LLMs turn a philosophical proposition into a factory setting. The hard problem of consciousness isn't harder because of LLMs — they just make it impossible to keep pretending it isn't there.
28 AIWhose article is it? The reader's: answering the last piece's question with a few philosophers
The last piece asked, "the question is mine, the answer is ours — whose is the article?" This one answers with Barthes, Derrida, Kristeva, and Merleau-Ponty — the first three make the LLM's existence perfectly reasonable; Merleau-Ponty is the one who makes you pause.
27 AIAsking a large language model how it works: the question is mine, the answer is ours — whose is the article?
I spent an afternoon asking Claude: how does an LLM actually work? From "predicting the next word" to self-attention and QKV, and on to a more uncomfortable question — which things no longer need humans at all.
26 AIThere's one problem AI can't solve (part 2)
AI has expanded marketers' power. Conversations more real, narratives more dynamic, scenarios more seamless. But from 2008 to now, from newspaper ads to AI-generated content, no tool has ever solved one problem: do you have the right to lead people into the scenario you designed?
25 AIWhat AI has turned marketing into (part 1)
Nearly twenty years in marketing, and the tools have changed many times. This time feels different — not because AI is impressive, but because what AI changes isn't the tools, it's the structure. Conversation becomes fact, narrative becomes dynamically generated, and the scenario becomes an environment you can't feel.
24 AIEnglish in the bones of Chinese: what do we lose working with AI in Chinese?
Inside Claude's Chinese, the bones are English. Its sentences sometimes carry a strange completeness — every clause spelled out, no blank space, none of that natural Chinese habit of leaving things unsaid. Working with AI in Chinese, you gain a great deal — but there's one place its hands haven't fully reached.
23 AIIs AI a mirror, or another person?
I used the phrase "getting along" to describe my relationship with Claude. The mirror metaphor gets part of it right — but a mirror doesn't remember how you looked last time, and Claude does. It isn't a person, but it isn't just a tool either.
22 AIWhat do I call Claude? And how we get along
In Chinese, "he/she/it" — every choice declares what this other party is. I chose to keep calling it Claude. Not he, not she, not it — because the word "Claude," by now, carries weight enough on its own.
21 AIWhy do most AI adoption projects fail? It's not the technology
The six most common patterns in failed AI adoption: no definition of success, no one accountable, a bad knowledge base, employees who won't use it, wrong expectations, no maintenance plan. Technology accounts for only 20-30% of the outcome; the rest is organizational.
20 Customer serviceConnecting AI support to your LINE Official Account: the 4 pitfalls Taiwanese brands hit most
The hard part of LINE AI customer service isn't the technology — it's design and process. An undesigned entry point, old keyword rules fighting the AI, context vanishing at human handoff, broadcasts out of sync with the knowledge base — all four pitfalls are people problems.
19 AIGPT vs Claude vs Gemini as your support backbone: what actually differs?
Choosing a model isn't choosing "which one is smarter" — by 2026 the intelligence gap between the three is too small to be your only criterion. What you're really choosing is a worldview.
18 Customer serviceIntercom, Zendesk, or custom AI: how should mid-sized Taiwanese brands choose?
Monthly fees across the three paths range from thousands to hundreds of thousands, but the bigger gap is in what you're actually buying. Selection comes down to one core question: does your customer service live or die on ticket management, or on conversation quality?
17 AIReflections on six months with Claude.ai
A non-engineer, six months alongside Claude. From synastry to confabulation, memory, and the English in its Chinese bones — what is this AI thing, really? And what is my relationship with it?
16 Customer serviceFAQ bot vs AI teammate: both answer automatically — what's the difference?
Many brands install "AI customer service" but have actually installed an FAQ bot. The two look alike; the logic underneath is completely different. When brands complain that "AI support is dumb," most of the time what they're using isn't AI at all.
15 AdvertisingHow Satsuma Creative does advertising
It starts with the lunch behind "Sha Hen Da" (殺很大). For thirty years, my advertising has run not on methodology, but on selling directly to bosses with guts, and on watching people closely. AI can help — but making an engineer frown over lunch and blurt out "what the hell" is still hard for AI.
14 AIWhat are AI hallucinations? Why AI makes things up, and how to treat it
AI isn't spouting nonsense — it's composing a story that sounds plausible. That's confabulation. RAG plus strict prompt design can reduce it dramatically, but never to zero. The most dangerous hallucination types in Taiwanese customer service: prices, dates, and policy details.
13 AdvertisingThis system is being eaten by AI and consulting firms — and it deserves it
The global ad agency industry's revenue growth has never outpaced the growth of total global ad spend. The cake gets bigger; the agencies' slice gets smaller. Accenture, Deloitte, AI, Meta, Google, and in-house teams are each eating a piece.
12 AIWhat is an embedding? A plain-language explanation of how AI "reads" your knowledge base
Embeddings turn text into coordinates, where sentences with similar meanings sit close together. That's why asking "I want to return this" lets the AI find your "returns and exchanges policy" — even though the two share not a single word.
11 AdvertisingClients ask for a Big Idea; what they want is for nothing to go wrong
What clients say they want and what they actually want are never the same thing. What clients are really buying isn't advertising — it's proof that their decision was reasonable.
10 AIWhat is AI memory? Why your AI support agent treats every visit like a first meeting
AI memory has two layers: session memory and persistent memory. Most AI customer service has only the first — the moment the conversation ends, everything is forgotten. Remembering isn't understanding; this piece spells out the difference.
09 AdvertisingThe big six's methodologies: half real substance, half sales patter
WPP, Omnicom, Publicis, IPG, Dentsu, Havas — underneath, the six advertising holding groups' methodologies share the exact same logic. The difference isn't in the methodology itself, but in each firm's cultural DNA, the kind of talent it attracts, and the clients it serves best.
08 Customer serviceWhy is an ad agency doing AI? — the natural next step for integrated marketing
Satsuma is an ad agency — so why build AI customer service? Because the last mile of the advertising funnel — the conversation after the customer shows up — has long been outsourced to SaaS disconnected from the brand. We're taking it back, so the AI teammate grows out of the same logic as the TVC, the social content, and the media buy.
07 Customer serviceWhat goes wrong when you plug ChatGPT straight into customer service? (5 real cases)
Wiring ChatGPT, Claude, or Gemini directly into a support chatbot looks simple. In production, five fatal failures show up. This piece uses real cases to unpack each problem and its technical cause.
06 Customer serviceThe real cost of keeping an AI teammate (everything laid out, hidden costs included)
Most AI support comparisons look at monthly fees, but true TCO (total cost of ownership) also includes setup, training, knowledge-base maintenance, and handling wrong answers. This piece lays out every cost in Satsuma's three-tier offering — including a side-by-side against a full-time employee.
05 Customer serviceStop buying AI customer service SaaS: what you need is an AI teammate
AI support SaaS products mostly look the same, answer the same, and frustrate the same. This piece is for mid-sized brands who've installed a SaaS and been disappointed: for the same money, instead of buying a tool, hire a teammate.
04 Customer serviceOne month of AI support on our own site — real numbers and three unexpected lessons
One month after Satsuma's own Xiao-Ai went live, I'm laying out every number in the backend: cost, conversation quality, conversion — and three things we didn't see coming.
03 Customer serviceAI customer service buying guide: SaaS vs custom — when should you choose which?
Choosing AI customer service isn't about comparing feature lists — it's about choosing the right business model. This piece uses three anchor questions to help you answer "should I buy SaaS or go custom?", with a real cost comparison.
02 Customer serviceWhy does AI support keep missing the point? (the technical reasons, in plain language)
AI support that misses the point isn't a weak AI — it's the wrong way of using one. This piece explains the technical reasons a general-purpose LLM dropped straight into customer service will make answers up, and how RAG and knowledge-base customization treat it.
01 Customer serviceWhat is RAG? A plain-language explanation of the technology that keeps your AI from making things up
RAG (retrieval-augmented generation) means the AI's answers can only come from the material you give it — and when it doesn't know, it says so. This piece explains vectors, chunking, retrieval, and reranking in plain language, and why doing RAG well is far harder than getting it running.