Over the past eighteen months, the term "AI-powered" has appeared in more than 2,000 product launches across consumer goods, enterprise software, and services. According to a review of investor materials and press releases, approximately 87% of these products made no meaningful change to their underlying technology. They changed the label. In many cases, that was enough.
The pattern has become so pervasive that analysts now have a name for it: AI washing. It is the practice of rebranding an existing product or service as artificial intelligence in order to attract investment, justify price increases, or simply remain visible in a market that has decided "AI" is the only word worth hearing.
The Rebrand Economy
In October 2025, a Denver-based mattress company called RestWell filed updated branding materials with its retail partners. The mattresses were the same. The foam density was the same. The coils had not changed. But the company was now RestWell AI, and its flagship product was called the "Adaptive Intelligence Sleep Platform." The marketing copy described how the mattress "learns your body" through "embedded micro-response technology." This technology, upon closer inspection, was memory foam. The kind that has existed since the 1970s.
RestWell's Series C round closed at $94 million, up from a $31 million Series B. The mattresses still cost $1,200. Customer satisfaction scores did not change. The investors were satisfied.
A regional sandwich chain in the mid-Atlantic rebranded its online ordering system as "machine-learning-curated condiment pairings." The system was an if-then logic tree built in 2019. If you ordered turkey, it suggested Swiss cheese. This is not machine learning. It is a dropdown menu with opinions. The chain's valuation tripled after the rebrand, and it secured a partnership with a food delivery platform that was specifically seeking "AI-native restaurant partners."
The Enterprise Playbook
Consumer products are the most visible examples, but the real volume is in enterprise software. A survey of 340 B2B SaaS companies found that 62% had added "AI" or "intelligent" to at least one product name in the past year. Among those, fewer than a quarter had introduced any new model, algorithm, or data pipeline. Most had added a chatbot to their help documentation. Some had integrated a third-party API that auto-generates email subject lines.
The pricing, however, changed substantially. Products rebranded as "AI-powered" saw an average price increase of 34%. The most common justification in sales materials was "advanced intelligence capabilities." When customers asked what those capabilities were, the most common answer, according to support ticket analysis, was some variation of "smart automation." When customers asked what that meant, the conversation typically ended.
One HR software company renamed its scheduling tool from "ShiftPlanner" to "ShiftPlanner AI." The product roadmap showed no engineering changes for the quarter in which the rename occurred. The only modification was a small sparkle icon added next to the logo. The company's next earnings call mentioned "AI" forty-three times. The previous quarter's call had mentioned it twice.
Who Benefits
The incentives are not complicated. Venture capital firms have made it clear, both publicly and in private term sheets, that AI companies receive higher multiples. A 2025 analysis of seed-stage funding found that startups describing themselves as "AI-first" received valuations 2.4 times higher than comparable companies that did not, controlling for revenue, team size, and market. The product did not need to use AI. The pitch deck did.
Public companies face a version of the same pressure. Earnings calls that mention artificial intelligence correlate with short-term stock price bumps, regardless of whether the company has shipped anything new. The word itself has become a financial instrument. It does not describe a technology. It describes an expectation.
The people who benefit most are the ones furthest from the product. Investors benefit from inflated valuations. Executives benefit from narratives that attract capital. Marketing teams benefit from a word that fits in every headline. The engineers, meanwhile, are often the ones fielding customer complaints about the "AI features" that do not exist, or that exist only as a tooltip that says "Powered by AI" above a search bar that works exactly the way it did before.
The Quiet Cost
There is a real cost to this, and it is not evenly distributed. When every product claims to be AI-powered, the companies that are actually building meaningful capabilities become harder to distinguish. A startup that spent three years training a proprietary model on medical imaging data gets listed alongside a company that added a ChatGPT wrapper to a contact form. They compete for the same investor attention, the same conference speaking slots, the same procurement budgets. The signal disappears into noise.
Customers, meanwhile, are learning to distrust the label. A recent consumer survey found that 71% of respondents said the phrase "AI-powered" made them less likely to trust a product, not more. The word has crossed the threshold from aspiration to skepticism in record time. It took "organic" about fifteen years to reach the same point. "AI" did it in three.
What Comes Next
Regulatory bodies in the EU and the FTC in the United States have both signaled interest in defining what "AI" means in product marketing, the same way "organic" and "natural" eventually required substantiation. Whether those definitions arrive before the next hype cycle remains an open question.
In the meantime, the rebrands continue. A commercial cleaning company in Ohio now describes its mops as "AI-enhanced surface optimization tools." The mops are the same mops. They have always been the same mops. But the proposal landed a county government contract last month, because the procurement form had a checkbox for "AI/ML capabilities," and someone was able to check it.
The checkbox did not ask for proof. It rarely does.