AI companies and established technology firms alike have seen products falter, shut down, or change course. Some could not find enough users or a sustainable business model. Others ran into competition, technical problems, privacy concerns, or legal pressure. Together, these cases show that an AI idea can attract attention without proving it belongs in a lasting stand-alone product.
When a separate product loses its place
Relay, an AI-powered workflow tool built as an alternative to Zapier, shut down entirely in August. It helped users automate email and task workflows with AI agents. But as OpenAI, Google, and other larger platforms added similar automation to their products, Relay struggled to maintain a distinct reason to exist on its own.
That pattern also appears inside larger companies. OpenAI’s attempt to recast ChatGPT as a broader “super app” brought separate “Chat,” “Codex,” and “Work” modes together, while renaming the familiar version “ChatGPT Classic.” Users criticized the redesign as confusing and cluttered, and OpenAI soon restored the familiar interface.
Other OpenAI products followed a different path: their features were absorbed into ChatGPT. ChatGPT Atlas, a stand-alone AI web browser, was discontinued on August 9 after less than a year, with its most useful features folded into ChatGPT. The company also shut down Operator, an agent that could browse the web and complete tasks, while image generation increasingly became part of ChatGPT, reducing DALL-E’s role as a separate destination.
These shifts suggest that a feature may continue even when its original product does not. For users, the experience can become simpler if useful capabilities sit in a service they already use. For the company, however, the challenge is deciding whether a stand-alone product brings enough distinct value to justify maintaining it.
Useful ideas still need demand and sustainable costs
Not every product has a larger platform waiting to absorb its features. Some close because the audience, economics, or performance never supports continued operation. According to S&P Global Market Intelligence, about 42% of AI initiatives are ultimately abandoned by their corporate parents. Reasons include insufficient funding, technical challenges, competition, difficulty scaling, and weak user demand.
Humane’s AI Pin tried to deliver AI through a wearable device instead of a conventional smartphone. The company raised $230 million and drew considerable attention, but the product struggled with performance. Humane also warned customers to stop using its charging case because of a potential battery fire risk. It shut down its AI Pin business in February 2025, and HP acquired most of the company’s assets for $116 million.
Rabbit R1 drew strong initial interest, with Rabbit saying it had sold 100,000 units shortly after launch. Early reviews, however, described an unfinished device with unreliable performance and few useful integrations. Rabbit has continued updating it, repositioning the R1 as a computer controller for agentic tasks, and announced a new hardware project called “Project Cyberdeck.” Its continued development shows that a rough launch does not always mean a company will abandon a product.
Other services encountered a more direct sustainability problem. Figgs AI let users create and interact with customizable characters for role-play and storytelling. Although the platform reportedly drew more than 1 million users, its developers said keeping the service free became too expensive. Yupp, which let people compare responses from hundreds of AI models and vote on them, shut down in March 2026 after its founders said it had not achieved strong enough product-market fit.
Privacy, legal pressure, and delayed promises
Some AI setbacks center on trust and the gap between a product’s promise and its readiness. Microsoft introduced Recall as a “photographic memory” for Windows PCs: it periodically captured screenshots so users could search their digital history. The announcement prompted concerns that such an archive could expose sensitive material, including passwords, private messages, and financial details.
Microsoft delayed Recall for nearly a year to reconsider its security and privacy protections. Even after a redesign, questions remained. A cybersecurity researcher built a tool to extract and display data Recall had captured, renewing doubts about whether the underlying risks had been addressed.
Apple’s improved Siri was a highlighted part of Apple Intelligence when the company unveiled it in 2024. The promised assistant would understand context, follow what was happening across apps, and carry out tasks. Apple repeatedly pushed back its release, citing engineering issues and bugs. The delay contributed to a $250 million settlement over claims about how Apple marketed the iPhone 16’s AI capabilities.
Siri AI finally appeared in the iOS 27 beta this past July. The new version is rolling out to English-language users first this month, with support for other languages expected to follow. The case illustrates how delays can affect both product expectations and the claims a company makes about what its technology can do.
What the shutdowns leave behind
Character.ai’s experience brings together several pressures. The consumer app let people chat with customizable bots, including characters modeled on celebrities and fictional figures. After raising $150 million at a $1 billion valuation in 2023, the startup saw Google pay a reported $2.7 billion the next year to license its technology and bring its founders back into its orbit.
Legal and regulatory challenges followed. Families sued, alleging the chatbots contributed to teen suicides and self-harm, while Disney sent a cease-and-desist letter over unauthorized use of its characters. Character.ai barred under-18 users from open-ended chats late last year and agreed to settle several suits, along with Google, in early 2026. The company filed for bankruptcy in early October, citing legal, regulatory, and operational headwinds. Its app remains available during the process, and Disney is the opening bidder with a $20 million offer for a license to some of its intellectual property.
Meanwhile, Notion Mail, launched in April 2025 to help organize and automate inboxes, is scheduled to shut down on September 22 after the company observed users relying on separate AI agents for email. Huxe, an AI audio app that turned written information into conversational, podcast-style audio, shut down in May 2026 as larger platforms began offering similar experiences to existing audiences.
These outcomes are not all the same: a product can disappear, be folded into another service, or keep operating with a new direction. But the recurring test is whether the product delivers reliable value that users want, while its makers can sustain it and address the risks it creates. The AI graveyard records missed bets, but it also shows which parts of those bets may carry forward.