Don't Ask the Robot What to Do


THE MACSPARKY DISPATCH

Hi Friend,

There is a world of difference between asking a robot, “Help me do this,” and asking, “What should I do?” AI can be remarkably good at the first question but lead you astray with the second. New users often learn that distinction the expensive way.

I’ve taught thousands of people how to use agentic AI, and I’ve seen a similar pattern. People get into trouble after they have some success with AI and try to push it further. Quite often, the thing they want on the other side just isn’t there.

At first, you’re skeptical of AI’s ability to provide any real help. With a little time and training, you figure out how to give it context, build skills, and assign the kind of donkey work it can reliably do unsupervised.

The moment that sinks in is liberating. Suddenly, you’ve got more time for creative work while the AI handles the grunt work.

As you get more reps in and the AI becomes reliable on this kind of work, your confidence in it grows. Then, somewhere along the way, you start giving it more credit than it’s due.

That’s where the drift begins.

The robot really isn’t equipped for the why. Put simply, the robot can’t exercise judgment and taste like you can. I’ve seen a lot of new models come out, and they keep getting smarter. Yet they still fail miserably in matters of taste and judgment.

I think the reason is context. The robot hasn’t lived your life. It doesn’t truly know your preferences. No matter how smart the model gets, its assertions of judgment are based on incomplete and inadequate context. It may have read a staggering amount of material, but it hasn’t been you.

I understand why people fall into this trap. The AI is fast, and it’s very confident in its answers. It feels authoritative. And yet, when you get beyond donkey work, these robots can be full of bullshit.

It doesn’t help that vendors of these frontier models keep promising the technology will change your life. Artificial intelligence is a transformational technology. It just isn’t nearly as transformational as the guy trying to pump up his company’s stock price claims.

People still give it more authority than they should. They ask it to do harder things than it can handle. After all that early success with donkey work, they start trying to make the AI smart enough to make those calls for them.

The fact is, it can’t make that judgment for you.

Trying to outsource judgment and taste to AI is an expensive way to get bad results. It’s like asking your toaster to pick the best bread for your turkey sandwich.

Also, even if it could exercise judgment and taste, do you really want it to? Isn’t that your job? Shouldn’t you get to decide what bread goes on your sandwich?

I see this often with people going through my coursework. They reach a point of frustration and give up on the utility of artificial intelligence because they pushed it beyond its capabilities.

My advice is always the same. Scale back. Get back to the useful zone.

Give the robot work where you know the outcome and can tell whether it’s any good. Limit its scope to jobs where you can provide enough context.

AI is great for grunt work that a computer can do. Stay in that space, and you’ll spend remarkably little time managing the AI. Draw clear lines in your head about how far you can take it and where it needs to step aside so humans can take over.

The reason I like the term “donkey work” so much is that it captures what AI does best. Granted, that doesn’t justify a trillion-dollar stock valuation, but it can make a real difference in your life.

You’re the human. You make the map. Hand the robot the shovel.

When you go beyond that, friends, there be dragons.

Your pal,
David
​

​Read this post on macsparky.com​


This Week in the MacSparky-Verse

Here’s what ran at MacSparky.com this week.

  • ​From Apple Watch to AI Assistant with Whisper Memos explains how I’m using the new read-only MCP connection to turn Apple Watch dictation into instructions for an AI assistant.
  • On Intentional AI 9, Chris and I separate real robot risks from extinction headlines and draw the line between robot drudgery and human judgment.
  • On Mac Power Users 867, Stephen and I compare notes on macOS 27, iOS 27, Siri AI, and the features that survived a summer of beta testing.

In the MacSparky Labs, we covered a lot:

  • I posted two short pieces on finding real savings in Apple’s refurb store and proving your backup can restore a file.
  • I made a hands-on video testing Siri AI with ordinary tasks involving calendars, messages, music, and visual intelligence.
  • I made a video setting up the Airversa QliQ button to control blinds and record scenes without waking the Mac.
  • We held a Deep Dive on turning a Mac Studio into a home server and a jam session about browsers.
  • I published The Lab Report podcast on the 27.2 betas, the new Mac mini, Vision Pro, and Mayday.

Next week, we’re getting together for our Q4 planning meetup.

That was a single week, and the Labs run like this every week. If that sounds like your kind of place, come join us.


This week's newsletter is sponsored by MailMaven, an email app built for Mac power users.

MailMaven lets you add project names and notes to your messages, so the context can stay with the email. Its conversation view brings together the messages in a discussion even when they're spread across mailboxes.

The automation is the part that interests me. MailMaven has an AppleScript dictionary for finding messages and working with them. SmallCubed also has an MCP server that lets a robot assistant search and manage the mail in the app. That gives you a way to connect your inbox to the work you're doing elsewhere on your Mac.

MailMaven is also fully compatible with macOS 27 Golden Gate, so moving to the new system doesn't mean leaving your email setup behind.

MailMaven's rapid maturation is truly impressive and if you're looking for a new email client, you'll want to check this one out.

If you've been looking for more control over your email workflow, take a look at MailMaven. My thanks to SmallCubed for supporting MacSparky.

22365 El Toro Rd, Unit 2168, Lake Forest, CA 92630
​Unsubscribe · Preferences​

David Sparks (MacSparky)

In a world where technology is increasingly conspiring to steal our focus and attention, my goal is to teach you how to be more productive with Apple technology. I want to help you achieve what is most important to you and enjoy your life at the same time using technology instead of becoming another one of its victims. Pretty much everything I make points at that North Star. I believe in this message so much that I’ve staked my livelihood on it.

Read more from David Sparks (MacSparky)

THE MACSPARKY DISPATCH Hi Friend, You may have noticed (or not) that I've been doing a bit more publicly on YouTube recently. I'm familiar with the video-making process. I've made over a thousand of them for all the Field Guides combined, and I'm approaching 400 videos in the MacSparky Labs. What I hadn't done much of, however, is public videos. So over the last several months, I've been making more of an effort there. I've got some editing help to give my videos a bit more of the public...

MacSparky Logo

THE MACSPARKY DISPATCH Hi Friend, This past week, Apple took the wraps off the new iPhone Pro and the brand-new iPhone Duo. It’s a big deal, and a lot of ink has already been spilled over it. But something else happened that I’d like to shed more light on. John Ternus gave his first keynote as Apple CEO. And his opening remarks in that job carry real weight. He chose to use that moment to make his case for Apple’s place in AI. (Yes, he called it AI.) It starts at the 3:32 mark in Apple’s...

MPU 865: Harvard Professor Teddy Svoronos on AI in the Classroom On this episode of Mac Power Users, Teddy Svoronos returns to compare the iPad and Mac in his teaching workflow, explain why he replaced Keynote with PowerPoint plus a custom AI-built remote, and share tools ranging from OpenIn and Macuse to LookAway and Wispr Flow. The conversation turns to the harder question of teaching in the age of AI: preserving judgment and brainstorming, rethinking take-home assessment, and checking what...