![](\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2025\/12\/9a649052175a0012fddadccf0125e533ec-screentimeairbnb-moshed.rhorizontal.w1100.jpg\")
<\/p>\n <\/p>\n
\n Photo-Illustration: Intelligencer; Photo: Airbnb\n <\/p>\n
One of the promises underwriting the AI boom is that software will be able to use computers on our behalf. At the low end, this looks like the \u201cdeep research\u201d tools AI firms have already released that search the web and synthesize information on your behalf, attempting to automate the task of using Google. More theoretically, it might mean models trained to use productivity software in a work context, which may then be used to attempt to automate increasingly complicated jobs. In between, you\u2019ve got the stuff that companies like Google and OpenAI keep touting in demos: AI that can book flights for you, AI that can get you a reservation, AI that can comparison-shop for you. As a category of \u201cagentic\u201d behavior, this sort of stuff is both ambitious and conceptually funny \u2014 the great big intelligence in the sky is being deployed, first, to click-farm the entire economy<\/a>.<\/p>\n So far, the results of such an approach have been mixed: Computer-use agents and self-clicking AI browsers make for great demos but struggle to offer obvious utility to most people. One might argue for a different path to broadly useful and flexible AI agents, one that doesn\u2019t take a long detour through web interfaces designed for human cursors and thumbs.<\/p>\n Or, as many in the AI industry would have it, you might argue instead that this is an obvious first step and that these tools don\u2019t yet work well only because they haven\u2019t been adequately trained and because they don\u2019t have enough access to the types of sites they\u2019re intended to use. That\u2019s how you get, as reported by Cade Metz at the New York Times<\/a>, AI start-ups making training-ready replicas of websites, including Gmail, Airbnb, and United Airlines:<\/p>\n These new shadow sites are a significant part of the tech industry\u2019s efforts to transform today\u2019s chatbots into A.I. agents, which are systems designed to book travel, schedule meetings, build bar charts and complete other computing tasks. In the coming years, many companies believe, A.I. agents will become increasingly sophisticated and could replace some white-collar workers.<\/p>\n \u201cWe want to build training environments that capture entire jobs that people do,\u201d said Robert Farlow, whose start-up, Plato, is among those recreating popular websites and other software applications.<\/p>\n The pitch for these companies is basically: We\u2019ll build replicas of popular websites so your AI can train on them and your agentic tools can attempt to use them with no limitations. Sending your half-baked AI agent out to try to book concert tickets will look, to the company selling them, a lot like spam, scraping, or sniping. Better, the thinking goes, to do most of this practice in simulation, where the airline might not even know you\u2019re planning to automate its ticketing interface.<\/p>\n There\u2019s a lot of similar stuff going on in AI, from \u201centerprise simulation,\u201d as Salesforce calls it \u2014 in which companies can use<\/a> \u201cvirtual environments that mirror the noise, accents, crosstalk, and complexity of real-world operations\u201d to train, for example, customer-service agents \u2014 to virtual environments where robotics companies can collect data from thousands of scenarios in gamelike environments that can then be used to train, for example, humanoid machines. Right now, on trade-show floors around the world, you\u2019ve got companies touting robots that have been \u201ctrained in simulation,\u201d often using the same software from Nvidia:<\/p>\n Say hi to\u00a0Dex!\u00a0 Our first\u00a0mobile humanoid robot\u00a0built for real-world industrial work.<\/p>\n Built for the factory floor, powered by\u00a0NVIDIA AI, trained in simulation, and ready for the real world.<\/p>\n Catch a first look of Dex at\u00a0#GTCDC25<\/a> Booth 368, where he\u2019ll show off his dexterity by\u2026 pic.twitter.com\/cU7wiHvgCh<\/a><\/p>\n \u2014 Richtech Robotics (@RichtechRobots) October 28, 2025<\/a><\/p>\n Like the software agents training on airline and e-commerce sites, there\u2019s an interesting tension at the core of humanoid-robot efforts: Does it make sense to automate tasks, or groups of like tasks, or to try to automate the entire person who typically does those tasks? But the singularity-in-a-sandbox parallel is illuminating in another way, too. Setting aside the question of how well such broad simulations map to the real world, the demand for such a thing is obvious. Running a bunch of tests with robot prototypes is prohibitively time-consuming and expensive, and doing the same thing in something akin to a video-game engine might get you to a productive feedback loop much more quickly. The reason you need a simulated Airbnb interface, though, is different: It\u2019s not that training a robot to use a website is particularly time or resource intensive. It\u2019s that Airbnb probably doesn\u2019t want you to build what you\u2019re building at all.<\/p>\n AI-training replicas are an early expression of a crucial question about the near future of AI agents, at least as companies like OpenAI have been talking about them: How will the internet, and the world in general, react to the presence of a bunch of freelance, mercenary machines? (Or, to put a finer point on it, to the replacement of human customers with bots transacting on their behalf?) Companies like Plato and AGI<\/a> exist because one of the more obvious ways for a company to train a computer-use agent, aside from monitoring users\u2019 screens or ingesting years of screen-captured videos \u2014\u00a0to have it try a whole bunch to see what works \u2014\u00a0is interpreted as antagonistic or threatening by the companies whose interfaces they\u2019re training on. Amazon is already suing<\/a> an automated-browser company for \u201cunauthorized access and trespass\u201d to its site, referring to a shopping agent as a sort of \u201cintruder.\u201d This is understandable but also revealing: To Amazon, customers are worth a lot more than the products they buy, particularly as audiences for advertising. Replica sites promise to give AI companies a means to get at least part of the way to a working product without interacting with companies like Amazon at all (although, as the Times notes, these companies are already getting sued too). Much in the way that the appeal of using ChatGPT instead of conventional search is partly because it just doesn\u2019t have many ads (yet), a chatbot shopping routine might be appealing because it lets you spend less time in Amazon\u2019s intentionally disorienting<\/a>, ad-laden, hypermonetized interface.<\/p>\n AI companies don\u2019t spend a lot of time dwelling, at least publicly, on what might happen after their agents-in-progress get deployed en masse. This is partly because they don\u2019t know. But the most obvious paths forward are all pretty bumpy. At one extreme, you have a situation in which every other company in the world has a severe immune response to what you\u2019re doing and sees AI agents as a generalized attempt to take over their relationship with their customers. At another, they happily accept that AI companies can provide them with a bunch of business, albeit under different terms than they\u2019re used to, and try to work something out.<\/p>\n Anything short of the latter scenario leaves AI firms with a fight. They could compromise, working out partnerships with, say, airlines, which, among other things, might mean that ChatGPT doesn\u2019t have to pretend to be a person using their websites but which would let airlines keep more control over how their tickets are sold through chatbots. But the rise of replica sites suggests that AI companies don\u2019t particularly want to ask for permission or partnership and that they know their approaches will be treated with suspicion. (Also, they\u2019re all building chatbot booking and shopping interfaces of their own<\/a>.) They\u2019d rather force the issue \u2014 see if these features work, see if users want them \u2014 then approach would-be partners with more leverage.<\/p>\n They\u2019re also a consequence of the AI industry\u2019s all-or-nothing ethos at the moment. Gradually incorporating a bunch of commerce partners into popular new chat interfaces is a plausible and familiar business plan, but at this stage of AI narrative cycle \u2014\u00a0and at this level of investment \u2014\u00a0plausible and familiar won\u2019t cut it. Partnering with Airbnb and getting commission on bookings probably doesn\u2019t add up to a trillion-dollar start-up valuation. Standing between your captive users and the entirety of the online economy, on the other hand, might. So that\u2019s what they\u2019re going to try, first in simulation, then in real life.<\/p>\n Sign Up for John Herrman column alerts<\/p>\n Get an email alert as soon as a new article publishes.<\/p>\n Vox Media, LLC Terms and Privacy Notice<\/p>\n