
Share
From Character.ai's bankruptcy to OpenAI's scrapped "super app," a growing list of shuttered AI products reveals a market correcting fast. Here's what the casualties tell us about where capital is misallocated.
The numbers tell a sobering story. According to S&P Global Market Intelligence, roughly 42% of corporate AI initiatives are ultimately abandoned by the companies that built them. That is not a fringe statistic. It is closer to a base rate, and it should recalibrate how investors think about risk in this sector.
Last month, chatbot platform Character.ai filed for bankruptcy, just five years after launching. A few weeks later, privacy-focused assistant Ollie announced its own shutdown. Neither was a surprise to anyone tracking the space closely. Both joined a lengthening list of products, startups, and corporate bets that failed to clear the bar: insufficient funding, technical limitations, brutal competition, scaling problems, or simply a market that never showed up.
Even the largest players are not immune. OpenAI, one of the most valuable private companies on the planet, has logged its own string of missteps. In July, the company tried to turn ChatGPT into a broader "super app," folding separate "Chat," "Codex," and "Work" modes into one interface while rebranding the familiar version "ChatGPT Classic." Users hated it. The redesign drew immediate criticism for being confusing and cluttered, and OpenAI rolled the change back soon after.
That was not an isolated incident. OpenAI has shut down or absorbed several stand-alone products into ChatGPT, a consolidation strategy that has worked better than the redesign did. ChatGPT Atlas, a stand-alone AI browser, lasted less than a year before being discontinued in August, its best features folded into the core app. Operator, an agent built to browse the web and complete tasks, got the same treatment. DALL-E has effectively been absorbed too, as OpenAI increasingly builds image generation directly into ChatGPT rather than maintaining it as a separate destination. Sora, the video-sharing platform, shut down in March after struggling with high operating costs and weak user retention.
The pattern across these failures is consistent, and it matters for anyone allocating capital to the sector. Stand-alone AI products are increasingly vulnerable to being out-featured by the platforms that already own distribution. Relay, an AI workflow automation tool positioned as a Zapier alternative, shut down entirely in August after five years in business. The company could not hold its ground once OpenAI, Google, and other large platforms built similar automation directly into their own tools. Notion Mail suffered a version of the same fate. Launched in April 2025 as an AI-driven inbox product, it is scheduled to shut down this fall after the company noticed users preferred routing their email through separate AI agents instead.
Huxe, an AI audio app built by former NotebookLM developers that turned written content into podcast-style audio, shut down in May, partly because larger platforms with existing audiences, Spotify among them, began rolling out comparable features. The lesson repeats: a well-built niche product competing against a platform with distribution and capital tends to lose, even when the underlying technology is sound.

Hardware has been even less forgiving. The Humane AI Pin raised $230 million and generated significant buzz before performance problems and a battery fire warning forced the company to shut down its AI Pin business in February 2025. HP acquired most of the assets for $116 million, a steep discount against the capital raised. The Rabbit R1, unveiled at CES in January 2024, reportedly sold 100,000 units quickly, but early reviews described it as unfinished, with unreliable performance and limited integrations. Rabbit has kept iterating rather than folding, repositioning the device as an agentic computer controller and recently announcing a new hardware project. Strong initial sales, in other words, did not guarantee a durable product, and the company is still working to prove the concept can last.
Not every story on this list is a clean failure. Apple's Siri overhaul, first promised in 2024 as part of Apple Intelligence, suffered repeated delays tied to engineering issues and bugs. The delay contributed to a $250 million settlement over how Apple marketed the iPhone 16's AI capabilities. Siri's new AI-powered version only reached the iOS 27 beta this past July, with a broader English-language rollout arriving this month. That is a costly, reputationally damaging delay, but not a graveyard entry, at least not yet.
Microsoft's Recall feature followed a similar arc. Pitched in 2024 as an AI-powered "photographic memory" for Windows PCs, it triggered immediate backlash over privacy and security, forcing a near year-long delay. Even after a redesign, a cybersecurity researcher recently built a tool capable of extracting data Recall had captured, reviving questions about whether the underlying problem was ever fully fixed.
Character.ai's unraveling is the starkest cautionary tale in the group. Founded in 2021 by former Google researchers Noam Shazeer and Daniel De Freitas, the company raised $150 million at a $1 billion valuation in 2023. Google then paid a reported $2.7 billion the following year to license its technology and bring the founders back in-house, leaving the startup to continue without them. Lawsuits followed, alleging the chatbots contributed to teen suicides and self-harm. Disney sent a cease-and-desist over unauthorized character use. Character.ai restricted under-18 users from open-ended chats and settled several suits alongside Google in early 2026. None of it was enough. The company filed for bankruptcy in October, citing legal, regulatory, and operational headwinds. Disney is now the opening bidder in the sale process, offering $20 million for a license to its intellectual property, with a resolution expected by late November or early December.
A 42% abandonment rate for corporate AI initiatives is not evidence that the technology does not work. It is evidence that the market is sorting winners from losers faster than in most prior tech cycles, and that distribution, capital reserves, and platform integration matter more than novelty. Investors should treat stand-alone AI products competing against incumbents with existing user bases as structurally higher risk, regardless of how compelling the underlying technology looks. The graveyard will keep growing. The question worth asking of any new entrant is not whether the product works, but whether it can survive the platform it is competing against eventually building the same feature for free.
Tags
Original Sources
The AI graveyard: a running list of projects and startups that didn't make it | TechCrunch
↗ https://techcrunch.com/2026/11/14/the-ai-graveyard-a-running-list-of-projects-and-startups-that-didnt-make-it
About the author
Marcus began tracking AI's market implications in 2016, noticing AI-related patent filings accelerating ahead of earnings upgrades before most of the sell-side had caught on. A former fixed-income quantitative analyst, he spent two decades building models that priced risk across emerging markets before pivoting to cover the economic impact of AI full-time. His writing translates opaque technical developments into clear risk/reward terms — and he's rarely diplomatic about the gap between AI valuations and underlying fundamentals. He believes most market participants still underestimate AI's long-run deflationary effect on knowledge work.
More from The Analyst →This Week's Edition
15 November 2026
14 articles
Related Articles

Tether Evo's New Research Tackles BCI's Calibration Problem with Cross-Subject Models
Models & Research · 5 min

Why Some Engineers Think the Ocean Floor Is the Future of Data Centers
Environment & Climate · 5 min

Europe's Digital ID Push Tests a New Privacy Tool: Zero-Knowledge Proofs
Policy & Regulation · 5 min
Related Articles

Tether Evo's New Research Tackles BCI's Calibration Problem with Cross-Subject Models
Models & Research · 5 min

Why Some Engineers Think the Ocean Floor Is the Future of Data Centers
Environment & Climate · 5 min

Europe's Digital ID Push Tests a New Privacy Tool: Zero-Knowledge Proofs
Policy & Regulation · 5 min
More Stories
© 2026 Cedar & Bloom. All rights reserved.