Actually Tested AI

How to Use Claude to Actually Understand a Long Boring Document Last week I had to read a 47-page municipal zoning variance report for a project I'm consulting on. The kind of document where every paragraph feels like it was written specifically to make you stop reading. I'd been putting it off for three days when I remembered: this is literally what Claude is supposed to be good at. The Obvious Approach That Didn't Work My first instinct was lazy. Copy the whole thing, paste it into Claude, type "summarize this." The summary I got back was technically accurate and completely useless. It told me the document was about a zoning variance request for a mixed-use development. Yeah, I knew that from the title. The problem with "summarize this" is that Claude doesn't know what you actually need from the document. It gives you the Wikipedia version — the facts without the judgment. For a 47-page zoning report, that meant I got a neat little ...

How to Use AI to Turn Meeting Notes Into Something People Actually Read Last Tuesday I watched my project manager paste raw meeting notes into Claude and get back something that looked like it was written by a corporate communications team having a bad day. Bullet points everywhere. Action items so generic they could apply to literally any meeting in history. The standard AI summary output that everyone ignores. I've spent the past three weeks testing different approaches to this exact problem. Not because I love meetings — I actively avoid them — but because the gap between "notes exist" and "people actually know what happened" is where most projects quietly die. The Default Output Is Useless (And Why) Here's what happens when you paste meeting notes into Claude with a basic "summarize this" prompt: you get a summary. Technically accurate. Completely ignorable. The AI extracts what was said without understanding what mattered. I ...

  So I've been poking at Google's AI search features for about eight months now, ever since the AI Mode numbers started looking genuinely surprising. At Google I/O three weeks ago, the company announced that AI Mode had crossed a billion monthly active users in its first year. That number stopped me. A billion. In one year. WhatsApp took five years to hit that. Instagram took about six. Google's AI search mode, which most people hadn't heard of eighteen months ago, hit that ceiling faster than almost anything in consumer technology history. I want to talk about what that actually means — not the version where everything is fine and AI is just a smarter search bar, and not the version where Google is about to disappear. Both of those readings miss what's actually interesting about where things are right now. What AI Mode Actually Does That Regular Search Doesn't If you haven't spent time with it, AI Mode in Google Search is fundamentally differen...

How to Use AI to Prepare for a Difficult Conversation at Work I had to tell my project lead that I couldn't meet a deadline we'd agreed on three weeks ago. Not because something went wrong — because I'd overcommitted and only realized it too late. The kind of conversation where you know you're in the wrong and there's no clever framing that changes that. So I did what I've been doing for months when I don't know how to start: I opened Claude and asked it to help me prepare. What happened next genuinely surprised me, and not in the way I expected. The Setup That Actually Works I've tried this with multiple AI tools. Claude consistently handles difficult conversation prep better than the others, and I think I finally understand why. It doesn't try to make you feel better. It tries to make the conversation go better. Those are different things. Here's the prompt structure I landed on after probably a dozen variations: The specif...

Four weeks ago I had seventeen browser tabs open and a six-week-old trip I still hadn't booked. Flight comparison sites, travel blogs from 2021, a Reddit thread that had devolved into an argument about whether Prague was "dead now." I'd been going in circles. That's when I closed everything, opened a single ChatGPT window, and typed: "I need you to actually help me plan this trip. I'll give you every constraint I have. Don't give me generic advice." Two weeks later I booked flights to Tbilisi, Georgia. Not the state. The country in the South Caucasus. I hadn't seriously considered it once in those six weeks of planning. The AI brought it up on the second message. Here's everything that happened — including the parts that didn't go well. The Opening Conversation That Set the Tone My first instinct was to ask ChatGPT to recommend a summer destination. But I'd made that mistake with meal planning before — asking an open-e...

Claude AI Prompts That Actually Save You Hours Every Week Last Tuesday I realized I'd been using Claude for eight months and still copying the same long prompt over and over. Every time I needed a client proposal, I'd dig through old chats, find the prompt that worked, paste it in, tweak it slightly, and hit enter. Then I stumbled on something that changed how I work with Claude entirely — and it's not what most guides will tell you. The System Prompt Trick Nobody Explains Properly You've probably heard people say "just give Claude context." That's useless advice. What actually matters is layered context — and here's what I mean by that. Instead of cramming everything into one massive prompt, I break my instructions into three parts. First, a role statement that's weirdly specific. Not "you are a helpful assistant" but something like "You are a freelance writer who's been doing B2B SaaS content for six years. Yo...

One small change.  Completely different results. I was three months into using Claude daily before I stumbled onto this. Had just finished a frustrating session — the kind where you're rewriting your prompt for the fourth time, trying to get Claude to stop giving you that generic, over-explained response. You know the one. Where it explains what a paragraph is before writing your paragraph. I almost closed the tab. But then I tried something different, and I genuinely haven't gone back since. The Problem Most People Don't Realize They Have Here's the thing about Claude — it's trained to be helpful. Like, aggressively helpful. Which sounds great until you realize that "helpful" often means padding everything with context you didn't ask for, qualifications you don't need, and that weird thing where it restates your question before answering it. I write content for clients. When I ask Claude to draft an email, I don't need a pream...

847 unread emails.  Four hours. One AI. Last Tuesday morning I opened my personal Gmail and saw 847 unread emails. Not spam — actual emails I'd let pile up over six months. Newsletters I swore I'd read, receipts I meant to file, random confirmations from services I'd forgotten I signed up for. I'd tried inbox zero methods before. I'd tried filters. I'd tried just... ignoring it. Nothing stuck. So I spent four hours that day figuring out how to make Claude actually useful for this problem — and I'm genuinely surprised by what worked. Why Claude Works Better Than You'd Think for Email Organization Here's the thing about using AI for inbox management: most people try to paste entire emails into Claude and ask it to "organize" them. That's not how this works. Claude can't access your inbox directly — there's no integration, no plugin, no magic button. But that limitation is actually what makes it useful. What Claude...