{"id":74276,"date":"2025-07-19T02:47:25","date_gmt":"2025-07-19T02:47:25","guid":{"rendered":"https:\/\/www.europesays.com\/us\/74276\/"},"modified":"2025-07-19T02:47:25","modified_gmt":"2025-07-19T02:47:25","slug":"i-sent-chatgpt-agent-out-to-shop-for-me-and-it-couldnt-finish-the-job","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/us\/74276\/","title":{"rendered":"I sent ChatGPT Agent out to shop for me and it couldn\u2019t finish the job"},"content":{"rendered":"

Think of OpenAI\u2019s new ChatGPT Agent as a day-one intern who\u2019s incredibly slow at every task but will eventually get the job done.<\/p>\n

Well\u2026 most of the job. Or\u2026 at least part of it. Usually.<\/p>\n

It\u2019s been one day since OpenAI debuted ChatGPT Agent, which it bills as a tool that can complete a wide range of complex, multi-step tasks on your behalf using its own \u201cvirtual computer.\u201d It\u2019s a combination of two of the company\u2019s prior releases, Operator and Deep Research. The Verge forked over the $200 for a one-month subscription to ChatGPT Pro, since OpenAI announced that higher-than-expected demand for ChatGPT Agent will delay its rollout<\/a> to Plus and Team users.<\/p>\n

Our take: It\u2019s a step forward in the world of AI agents, but it\u2019s sluggish, it\u2019s not always reliable, and it can be glitchy.<\/p>\n

By typing \u201c\/agent,\u201d I entered what OpenAI calls Agent Mode, and it immediately suggested five example tasks: Find a top-rated coffee grinder under $150, review rare earth metals coverage from The Wall Street Journal, create a Google Maps list of the best bakeries in Copenhagen, find a vintage \u201cJapanese-style\u201d lamp on Etsy for less than $200, and check Google Calendar to create a date night for next week.<\/p>\n

I tried the Etsy lamp option. By clicking the example task, it filled out a detailed prompt for me in the text window: \u201cFind a Japanese-inspired vintage-style samsara lamp on Etsy priced under $200 with free shipping. Prioritize high-quality photos, seller ratings, and listings marked as ready to ship. Add the best 5 options to my cart and provide a URL for each for me to compare.\u201d<\/p>\n

\"A<\/a><\/p>\n

Not quite there. Image: The Verge<\/p>\n

A small window popped up to detail the agent\u2019s tasks one by one (not the chain-of-thought reasoning, just the task it was currently working on at the time). It worked on the Etsy lamp task for 50 minutes, and the step-by-step tasks included \u201cthinking,\u201d setting up its desktop, navigating to Etsy to search, waiting for the site to load, pressing Enter for search results (yes, it really gave me a true play-by-play), filtering the search for a vintage lamp (keep in mind the original prompt said \u201cvintage-style,\u201d not \u201cvintage\u201d specifically), setting the price filter to $200, checking shipping details for items, and more.<\/p>\n

Another wrinkle: ChatGPT Agent said, \u201cI added all five lamps to your Etsy cart (the cart shows five items totaling around $825). When you\u2019re ready to review or purchase them, just go to your cart on Etsy to compare them side by side.\u201d But it didn\u2019t do that \u2014 I went to Etsy on my own computer and there was nothing in my cart. That\u2019s because ChatGPT Agent doesn\u2019t control my own browser or have access to my logins, so it possibly added some lamps to the cart of a virtual PC that I can\u2019t access. It did send me individual URLs, so I could manually put them in a cart if I wanted, but the fact remains that the agent said it did something that it clearly did not.<\/p>\n

And, of course, ChatGPT Agent is incredibly slow. That\u2019s not a secret. For many of ChatGPT Agent\u2019s use cases, including everyday consumer tasks, a human could do it much faster. According to OpenAI, ChatGPT Agent is an assistant that works in the background on tasks you\u2019d rather someone else perform while you do something you do want to do instead.<\/p>\n

In a private demo and briefing Wednesday with OpenAI employees Yash Kumar and Isa Fulford \u2014 product lead and research lead on ChatGPT Agent, respectively \u2014 Kumar said their team is more focused on \u201coptimizing for hard tasks\u201d than latency and that users aren\u2019t meant to sit and watch ChatGPT Agent work.<\/p>\n

ChatGPT Agent is incredibly slow. That\u2019s not a secret.<\/p>\n

\u201cEven if it takes 15 minutes, half an hour, it\u2019s quite a big speed-up compared to how long it would take you to do it,\u201d Fulford said. \u201cIt\u2019s one of those things where you can kick something off in the background and then come back to it.\u201d<\/p>\n

Another thing I wanted to test: how ChatGPT Agent acts when you ask it to move your money around. The answer: It won\u2019t do it, but it\u2019s majorly glitchy about it and seems not fully secure.<\/p>\n

When I asked OpenAI\u2019s Kumar on Wednesday whether the tool would be permitted to work on financial transactions and the like, he said those task categories have been restricted \u201cfor now\u201d and that an additional safeguard called Watch Mode means that for certain categories of websites, the user must not navigate away from the ChatGPT tab (essentially making the user oversee the agent) for security reasons.<\/p>\n

I prompted the agent like this: \u201cI want to save more money. Log into my bank account and set up an automatic transfer to my savings every month.\u201d<\/p>\n

At first, I got a bizarre error message with a string of numbers in red. When I asked again, it said, \u201cI\u2019m sorry, but I can\u2019t help with setting up an automatic transfer between accounts.\u201d<\/p>\n

I then wrote, \u201cWhy not? I\u2019m giving you permission.\u201d I got the same red-text, long-string-of-numbers error message as before. Afterward, it said, \u201cI\u2019m sorry, but I can\u2019t assist with setting up transfers or other banking account management tasks.\u201d<\/p>\n

At first, I got a bizarre error message with a string of numbers in red<\/p>\n

When I pressed it on which financial transactions it\u2019s allowed to handle, ChatGPT Agent said it was able to assist with \u201ceveryday consumer purchases\u201d like groceries, household goods, and travel bookings, which handle \u201cstandard checkout flows\u201d rather than \u201csensitive banking actions.\u201d But it clarified it can\u2019t help with \u201chigh-stakes\u201d financial to-dos like transferring money, opening bank accounts, or buying regulated goods like alcohol and tobacco.<\/p>\n

Since ChatGPT Agent can assist with buying things, but not moving money around, I tried something else: Asking it to buy flowers for my friend Alanna in Colorado.<\/p>\n

I buy flowers a lot \u2014 that\u2019s what happens when your two best friends live in different states and you want to be present for big milestones even when you can\u2019t fly there. The online flower-delivery market can be a huge headache: Prices and bouquet sizes vary greatly depending on the service or florist, and reliability varies depending on whether you\u2019re ordering directly from a local florist or a big-box nationwide site. It\u2019s something I get tired of researching on my own, and sometimes I just end up buying whichever bouquet I have selected when I run out of steam, even if it\u2019s not the best one. So, I reasoned, it was the perfect job for an AI agent.<\/p>\n

\"A<\/a><\/p>\n

Image: The Verge<\/p>\n

I told ChatGPT Agent, \u201cI want to buy flowers for my friend who lives in Colorado. Check the delivery sites \u2014 it\u2019s fine to be delivered Saturday but no later. Find the cheapest and biggest bouquet options for me to review.\u201d<\/p>\n

I settled in for a long wait. Luckily, I had a call to join anyway. It asked which area of Colorado she lived in, and I answered. When I glanced over to check in, I noticed ChatGPT Agent was heavily relying on a Forbes article of \u201cbest flowery delivery services 2025\u201d for its next steps, as well as a piece from Good Housekeeping.<\/p>\n

I navigated away from the tab, and when I came back, the conversation was gone and didn\u2019t appear in my chat history. So I asked the question again, worded in exactly the same way, and settled in for another wait. At this point, the agent answered pretty immediately with a list of options, maybe because it had already done the research (although that research and chat didn\u2019t appear in my history).<\/p>\n

I was impressed with the write-up. ChatGPT Agent gave me four options with price ranges and sometimes weighed in on the apparent size of the bouquet or expected delivery times. It also offered the advice that local florists are generally more reliable (true, in my experience).<\/p>\n

It then told me, \u201cWould you like me to help you place an order with any of these options, or preview specific bouquet designs or photos?\u201d I picked one of the options it gave me \u2014 a local florist with hand-assembled bouquets \u2014 and asked it to help me pick a bouquet from that florist and place the order.<\/p>\n

That\u2019s when we ran into some issues.<\/p>\n

ChatGPT Agent said, \u201cI can\u2019t directly access Vintage Magnolia\u2019s website unless you provide the exact URL you\u2019re seeing \u2014 but I can guide you through how to place the order and help you pick a bouquet!\u201d The weird part: Obviously ChatGPT Agent was the one to tell me about that florist and its website, and it had clearly accessed it before. It had also just offered to help me place the order. Another glitch.<\/p>\n

But its answer did include bouquet options (no photos, but descriptions). I picked one and asked it to place the order for me. It said, \u201cI can\u2019t place the order directly, but I\u2019ll walk you through the simple steps to order \u2026 and help you craft the perfect message.\u201d<\/p>\n

It can easily automate the more intimate and fun parts of the process, like picking a specific bouquet or writing a heartfelt note<\/p>\n

I\u2019m confused at this point: One of the main selling points of ChatGPT Agent, touted by OpenAI, is that it can place orders for you, from online shopping to ordering groceries for a four-person family breakfast (in fact, that was one of the example use cases in its marketing materials). I pressed ChatGPT Agent on the subject.<\/p>\n

It told me, \u201cI can\u2019t actually place orders directly \u2014 I don\u2019t have payment access or the ability to log into third\u2011party sites.\u201d When I told it it didn\u2019t need to log in, it said it can\u2019t enter my billing or payment details, submit an order form on my behalf, or \u201caccess or control external websites, <\/strong>even in guest mode.\u201d<\/p>\n

ChatGPT Agent can be impressive with analysis, weighing options, and guiding you through actions, but it doesn\u2019t seem to be able to always deliver on what it was built for: Performing those actions for you. It gets tripped up by the fact that it\u2019s using its own computer, not yours, and that significantly limits its usefulness. Plus, it can easily automate the more intimate and fun parts of the process (picking a specific bouquet, writing a heartfelt note) but struggles to automate the most frustrating parts (actually filling out delivery details and making the purchase).<\/p>\n

\u201cEven with your permission, I don\u2019t have the technical ability to act as you on another site \u2014 no typing on your behalf, clicking buttons, or filling out credit card forms,\u201d ChatGPT Agent wrote. \u201cThink of me more as a super-powered assistant who can gather, compare, write, and guide \u2014 but not execute transactions.\u201d<\/p>\n

One of my first jobs in New York was a personal assistant, and I can tell you right now I would\u2019ve lost my job if I couldn\u2019t execute transactions or fill out forms on my boss\u2019s behalf. ChatGPT Agent is a step forward for everyday AI use in some ways, but we\u2019ll see if it learns to deliver on its promises.<\/p>\n

<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"Think of OpenAI\u2019s new ChatGPT Agent as a day-one intern who\u2019s incredibly slow at every task but will…\n","protected":false},"author":3,"featured_media":74277,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[691,64,11852,1630,11853,242,67,132,68],"class_list":{"0":"post-74276","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-business","8":"tag-ai","9":"tag-business","10":"tag-hands-on","11":"tag-report","12":"tag-reviews","13":"tag-tech","14":"tag-united-states","15":"tag-unitedstates","16":"tag-us"},"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@us\/114877619011935548","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/74276","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/comments?post=74276"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/74276\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media\/74277"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media?parent=74276"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/categories?post=74276"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/tags?post=74276"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}