A startup promised that their AI assistant would build you an app. But the work was actually done by human engineers.
Builder.ai, a startup backed by Microsoft, pitched itself as an AI-powered way to simplify app development. Clients chatted with the platform's signature AI assistant, Natasha, and received a functional, AI-generated app based on the information they provided. But instead of using AI technology to run the chatbot and create the app, the company hired 700 engineers in India to pose as Natasha in conversations with clients, and then to do the actual coding of the app.
The company's human-run chatbot operation is part of a larger problem in the tech industry today: An issue called "AI-washing," when tech companies purport that their tools use AI a far greater amount than they actually do. It happens remarkably often, like when Coca‑Cola claimed their 2023 product Y3000 Zero Sugar was co-created with AI, but provided no details on how AI was actually involved in the creation of the product, leaving many to speculate that the claim was designed to get more attention and interest from consumers.
As companies scramble to incorporate AI into their offerings — or at least, give the impression that they have done so — consumers may not share the tech sector's unfettered enthusiasm for AI everything.
The Pew Research Center reports that 43 percent of respondents think AI will harm them, in comparison to just 24 percent who think the tech will benefit them. Moreover, "Public optimism is low regarding AI’s impact on work," the Pew report reads. "While 73 [percent]of AI experts surveyed say AI will have a very or somewhat positive impact on how people do their jobs over the next 20 years, that share drops to 23 [percent] among U.S. adults." According to another study, about half of all respondents said they’d rather speak to a real person over AI, in comparison with just 12 percent of respondents who said they preferred to speak with an AI chatbot. A quarter of respondents said it depended on the situation.
But AI washing wasn’t the problem that got Builder.ai in trouble. According to the Latin Times, a lender seized $37 million from the company after discovering it generated just $50 million in revenue — 300 percent lower than its $220 million claim. Linas Beliūnas of Zero Hash accused Builder.ai of fraud in a LinkedIn post, writing: "It turns out the company had no AI and instead was just a group of Indian developers pretending to write code as AI." A former employee sued the company, Business Standard reported. An audit seized millions from the company. Now, it owes Amazon $85 million and Microsoft $30 million for cloud services it never paid for.
The company filed for bankruptcy in the UK, India, and the U.S. In statement on LinkedIn, Builder.ai wrote that it would be "entering into insolvency proceedings and will appoint an administrator to manage the company’s affairs."
"Despite the tireless efforts of our current team and exploring every possible option, the business has been unable to recover from historic challenges and past decisions that placed significant strain on its financial position," the LinkedIn post read.
