How to use ChatGPT agent to compare 14-inch laptops
Atlas Agent Mode compared Samsung and Apple 14-inch laptops using official pages; a Consumer Reports test checked accuracy, source clarity and browser friction.
consumerreports.org
Original prompt
Shared by the source author. See source notes for context.
Open tabs for the latest 14-inch versions of the Samsung GalaxyBook 5 Pro and Apple MacBook Pro, then create a table
Reach people exploring AI agents
See placements →What happened
The reviewer found Atlas’s table accurate in this test, while other browser runs exposed uncertain sourcing and permission friction; no laptop was bought.
Reported by Nicholas De Leon — Consumer ReportsHow the agent did it
- De Leon gave the browsers the same two-model comparison brief and requested price, battery and port information.
- Atlas Agent Mode consulted official pages and a secondary technical cross-check to create its table.
- The reviewer compared accuracy, source clarity and observed timing against Comet and Edge Copilot.
Where a person stepped in
De Leon evaluated the actual tables and their source provenance; the experiment did not delegate a purchase or independently benchmark general browser reliability.
Nicholas De Leon tested AI browsers on the same shopping-research task: compare the latest fourteen-inch Samsung Galaxy Book5 Pro and Apple MacBook Pro. He wanted a table covering prices, battery life and ports, with the products grounded in current source pages at the time of the test.
Atlas Agent Mode opened official product pages and cross-checked information with Laptop Mag. De Leon judged the resulting comparison accurate and reports a run of two minutes and five seconds. This was one observed task, not a general performance benchmark or an actual laptop purchase.
Two Comet attempts were less clear about source quality. The first said it could not reach official pages and relied on other information; the second presented official sources, but the reviewer still found the provenance uncertain. Edge Copilot completed its comparison more slowly and requested permission to use Samsung’s website.
The account shows why a polished comparison table is not sufficient evidence by itself. The human checked source selection and correctness across runs. The demonstrated outcome is useful product research from Atlas, with measured timing limited to the author’s experiment and inconsistent sourcing visible in alternatives.
Prompt
Original prompt
Source: Consumer Reports · Nicholas De Leon
Frequently asked questions
Which agent produced the strongest reported table?
The reviewer judged the Atlas Agent Mode output accurate for this particular comparison.
Did the test prove all Comet answers were wrong?
No. The criticism concerns inconsistent access and uncertain source provenance across the observed attempts.
Was Edge Copilot the coding product?
No. This comparison used the Edge browser’s Copilot, not GitHub Copilot.
More details
About this account
De Leon gave the browsers the same two-model comparison brief and requested price, battery and port information. Atlas Agent Mode consulted official pages and a secondary technical cross-check to create its table.
Full original primary HTTP HTML read and date/byline verified. Original prompt excerpt. Atlas maps to ChatGPT agent; Edge Copilot is not GitHub Copilot. Single-test timings are not generalized performance claims. Original prompt excerpt.
Credit: Nicholas De Leon — Consumer Reports
Source published 10 Nov 2025
Source checked 14 Oct 2026
Tools and topics
Atlas Agent Mode · Official product websites · Laptop Mag · Perplexity Comet · Edge Copilot
laptop comparison · source verification · browser evaluation
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.