Stop Asking AI Questions. Give It a Chore.

TL;DR: A new generation of AI assistants has crossed a threshold that changes everything. They no longer just answer questions. They take action on your behalf, booking dinner, filing the return, drafting the follow-up, all without you opening a single app. Here is what they are, which one fits your life, and the simple rule for handing over your first chore without regret.

This morning, while I was still on my first cup of coffee, my assistant texted me about a paid consulting invite that had slipped into my inbox.

I hadn't asked it to look. It just knew I'd want to know.

I didn't open a single app. I just read my texts.

That's the shift nobody is explaining very well. For three years, AI has been a brilliant conversationalist. You ask, it answers, and then the actual work is still yours. You write the email. You book the flight. You sit on hold.

The new assistants take the job, not just the question. Tell one "find us a dinner spot for Saturday and book it," and it goes and does it, often without needing you to lift a finger.

That's a bigger line to cross than it sounds.

I've Been Waiting Fifty Years for This

When I was twelve, my dad brought home a bag of parts. Inside was an RCA Cosmac VIP computer. After I finished soldering the circuit board together, I discovered the price of admission. To make it do anything at all, I had to type hexadecimal codes into a keypad. Literal ones and zeros, in the exact perfect order, just to move a single dot across a screen.

I was completely enthralled. And completely enslaved to the machine's language.

For the next fifty years, that was the deal we made with technology. You want to use it? Learn its grammar. Click its buttons. Organize your files exactly the way it demands. We became assistants to our assistants.

That era is over. And a machine that finally takes on tasks, instead of waiting for instructions it can barely parse, is the thing that closes it.

But before you hand anything over, let me put this moment in its proper context.

Been Here, Afraid of That

In 1865, Britain passed a law requiring someone to walk in front of every "horseless carriage" waving a red flag. No, seriously:

"Firstly, at least Three Persons shall be employed to drive or conduct such Locomotive, and if more than Two Waggons or Carriages be attached thereto, an additional Person shall be employed, who shall take charge of such Waggons or Carriages.

Secondly, one of such Persons, while any Locomotive is in Motion, shall precede such Locomotive on Foot by not less than Sixty Yards, and shall carry a Red Flag constantly displayed, and shall warn the Riders and Drivers of Horses of the Approach of such Locomotives, and shall signal the Driver thereof when it shall be necessary to stop, and shall assist Horses, and Carriages drawn by Horses, passing the same."

"...It shall not be lawful to drive any such Locomotive along any Turnpike Road or public Highway at a greater Speed than Four Miles an Hour, or through any City, Town, or Village at a greater Speed than Two Miles an Hour...."

So the very first "automated locomotives" (later simply known as automobiles) could transport people only as fast as a human could walk. People weren't wrong to be careful. They were just early. A generation later, nobody walked in front of cars, and nobody wanted their horse back.

Elevators went through the same thing. For decades, every elevator had a human operator. When automatic ones showed up, plenty of people refused to ride alone. What won them over wasn't a better motor. It was a big red stop button, an emergency phone, and a little time.

That's exactly where AI assistants are today: the automatic elevator, with the stop button built in.

What's Actually Under the Hood

Every one of these new assistants has four working parts:

• A brain: the same kind of AI you may have already chatted with. It reads, thinks and writes. You've been living with this piece for a while.
• Hands: this is the new part. It gets its own computer and can click, type and fill in forms just like you would.
• A memory: it learns how you like things done, like the barista who starts your order when you walk in.
• A leash: the good ones ask before they send, spend, or book anything with real consequences. Think of it as the valet key that starts the car but won't open the trunk.

No code. No magic words. You text it the way you'd text a capable friend.

Think of these as always-on assistants with their own computer, connections to your apps and a memory for how you like things done. Less hotel information desk clerks and more concierges who can actually make your arrangements.

Eight Assistants That Actually Do the Work

Here's where things stand in the fall of 2026. Different tools for different lives.

• Siri AI (Apple) is the one already in your pocket. I spent a month with the iOS 27 beta, and Siri finally works. It's free, in beta, English first. It can act on whatever is on your screen and take actions inside your apps. You'll need an iPhone 15 Pro or newer.
• Muse (Meta) the gentlest start. It lives inside WhatsApp or in a Mac app and it feels like texting a person. It asks before it sends an email or buys anything. Free to start, US only.
• Dots (OpenAI) is ChatGPT with a pair of hands. In OpenAI's demo, a Dot noticed a calendar packed through dinner and suggested delivery options, then waited for the person's choice and timing. While a cute animated blob is just the wrapping, the useful part is handing off the chore, with clear rules for when it needs your approval first. Only available on Pro $200 plan or higher
• Gemini Spark (Google) is built for Gmail households. It works around the clock across Gmail, Calendar, Drive, and Maps, even with your laptop closed. The catch: for now it's only available on Google's top paid AI Ultra plan.
• Grok Bot (xAI) is the tireless coworker. Each Bot gets its own computer, signs into your work tools and grinds through follow-up emails and invoices 24/7, checking in when it needs your OK. It's in beta, on paid SuperGrok plans.
• Perplexity Computer is the research department. Hand it a big job and it splits the work among a team of AI helpers who go off and report back, sometimes hours later. Plans run $20 to $200 a month.
• Instinct: the assistant you just text. It's the one I use every day. You text or call it like a person, and once you connect your email, calendar and messages, it follows up on threads you dropped, books the handyman, and arranges the ride to the airport. For now it's free but invite-only: join the waitlist or ask a member (like me!)
• OpenClaw is the kit car. Free, open source, and it runs on your own computer, chatting with you through WhatsApp or iMessage. Your data never leaves your house. Setup takes a weekend afternoon. Think Heathkit radio in the '70s: more work, but you own every part.

Which One Is Yours?

Pick the one that fits your life, not the one with the most buzz.

• Have a newer iPhone? Start with Siri. It's free and it's already there.
• Live on WhatsApp or Mac? Muse.
• Have tons of stuff already in ChatGPT's memory. Use Dots.
• Your life runs on Gmail? Gemini Spark, once it reaches your plan.
• Want one assistant for all the loose ends in your life? Instinct.
• Run a small business or do a lot of research? Grok Bot or Perplexity Computer.
• Like to tinker? OpenClaw.

This is AI for the Rest of Us, not just for engineers and early adopters. There is now a version for every kind of person on that list.

Break It In Like a New Babysitter

You wouldn't hand a new sitter your car keys and credit card on night one. First night, it's dinner and bedtime. Same idea here.

Pick one chore that bugs you, like the return you haven't shipped. Let the assistant draft, and you hit send. Don't pay a dollar until you can name what the upgrade buys you. Then hand over more as it earns it. Why?

These assistants stumble. So did early cars, which is why drivers carried a toolbox and a spare everything. Reuters reported that Meta's own testers watched Muse stall on tasks before launch.

My rule is simple: check the important stuff, especially at the start. A good assistant gets me 80% of the way there in 5% of the time. The last 20%, the judgment and the final read before something goes out with my name on it, is still my job.

I get it. Letting software act for you feels strange. So did stepping into an elevator with nobody at the controls. And things are changing so rapidly that what required a Mac Mini running OpenClaw in May is now a few chat messages away with Instinct. What a time to be alive and in tech!

So what's the first chore you're handing off?

About the Author

Steve Chazin makes AI make sense. After three decades leading tech teams at companies like Apple and Salesforce, he's on a mission to show regular people how to use AI without fear or confusion. No hype. No doom. Welcome to the Digital RenAIssance.

If this helped, subscribe at stevechazin.com for weekly updates. And if you know someone who's trying to make sense of AI, forward this to them. We're all learning this together.