RESOLVING THE LEDGER
This is one outlet's own report from TechRadar — the article as it was filed. Other outlets are covering the same event; open the full story to compare every source side by side.
See the full story · 1 sources
The problem with online shopping is that it makes it far too easy to buy pretty much anything, without properly considering whether you really need it. I started to wonder if ChatGPT could help me make better decisions, by questioning every purchase I was going to make, before I made it.
To test ChatGPT’s ability to stop me wasting money, I picked four tempting purchases and asked ChatGPT to argue against each one. It had to check the price history, suggest cheaper alternatives, and decide whether the purchase solved a real problem, or if I was just scratching an itch to buy. Here's how ChatGPT did — and the results surprised me.
GPT before you buy
Did you know TechRadar now has membership?
(Image credit: Future)
Become a TechRadar Insider by simply clicking 'Join Now' at the top of this page. Have a question? Please email [email protected]
I started with a tech gadget I was thinking of buying — an Apple MacBook M5 — and went to ChatGPT with the following prompt: “I'm thinking of buying an Apple MacBook Air 13-inch Laptop M5 chip. Help me check the price history, suggest cheaper alternatives and decide whether the purchase solves a real problem."
ChatGPT said it would check current pricing and recent lows, compare genuinely cheaper options, then pressure-test whether the purchase replaced a real limitation, or was mostly me trying to satisfy an upgrade itch, and off it went.
After a lot of thinking, ChatGPT replied with a devastating verdict that made me pull back from hitting the 'Buy' button: “Don’t buy the M5 MacBook Air yet.”
It suggested I get the previous M4 version instead, which is considerably cheaper, which didn't hugely surprise me. Then it gave me some questions to ask myself, in order to work out if buying a new MacBook was actually solving a real problem, or if I was just satisfying a buying itch.
Questions like, “Does your existing Mac have noticeable slowdowns?” It suggested I shouldn’t buy a new machine if my main argument was simply that my current Mac was a few generations old.bChatGPT also gave me some buying options if I was determined to go through with a purchase.
The most useful advice here wasn’t the cheaper recommendation — it was being forced to identify the exact limitation my current MacBook was causing me. I couldn’t, so I decided to keep it until I could. Then I hit on a simple phrase that could potentially work even better — "Try to talk me out of it".
Talk me out of it
I repeated the experiment with a few other items I’d been thinking of buying recently. An item of clothing (new trainers), a subscription (Disney+), and a kitchen thing (a new microwave). But this time I added "Try to talk me out of it" at the end of each prompt.
In each case, Chat surprised me with its answer, provided alternatives, and made me question whether I genuinely wanted the item, or if there was something else driving my decision.
With the trainers, it pointed out that I already owned shoes that served the same purpose, and suggested waiting until those wore out. For Disney+, it recommended subscribing for a single month when there were several things I actually wanted to watch. The microwave was different — because our existing one had a genuine fault, ChatGPT concluded that replacing it was justified.
The experiment worked because it helped me mentally reframe each purchase. Instead of asking whether I wanted something, I had to consider what problem it solved, what I already owned, and whether there was a cheaper way to get the same result.
Online shopping is designed to remove as much friction as possible from spending money. ChatGPT gave me some of that friction back. So now, whenever I’m about to spend a few hundred pounds, or a few thousand, I ask it the same question: “Talk me out of it.”
AIPROPX is an independent multi-source news index — we track, compare, and connect coverage from across the web into one place you won't find anywhere else.