Planet Fitness Can Manage Potential Risks to Churn, Membership Growth From Meta AI Agent, UBS Says
MT newswire2026/10/09 18:2202:22 PM EDT, 10/09/2026 (MT Newswires) -- Planet Fitness (PLNT) investors are concerned regarding potential implications on churn and membership growth from Meta Platforms' (META) new artificial intelligence agent, though the risk is "manageable in practice," UBS Securities said Friday. Last month, Meta said its Muse personal AI agent can perform tasks such as sending e-mails, booking travel and negotiating bills on behalf of users. The agent can also continue working after users close the app. "(Planet Fitness) sentiment has been negative and took a worse turn after Muse launch and fears that agentic AI can drive significant impact on churn in the medium-term," UBS analysts, including Arpine Kocharyan, said in a note to clients Friday. "We believe agentic AI risk is sensible but manageable in practice." The fitness center operator could introduce authentication steps, such as in-app verifications and biometric confirmations, which AI agents would struggle to complete, according to the brokerage. Planet Fitness could also restrict automated access, block bot traffic at the network level, and require direct confirmation from members before allowing an AI agent to cancel their subscriptions, UBS said. The high cost of rejoining gyms could keep members to not allow automatic cancellations, while Planet Fitness can use the AI technology itself to personalize outreach and offer membership freezes or downgrades before a cancellation, according to the note. Planet Fitness' business fundamentals have improved amid stronger membership growth in September and continued strength in early October, UBS said. "Churn has been stable in the 3.5%-3.8% range, with (Planet Fitness) likely implementing a change in billing whereby a member is billed same day (versus) 30 days later," the analysts said. "If implemented, this could further reduce churn." More stable churn means Black Card pricing hike could come "sooner than the market believes," the analysts wrote. Black Card membership typically has higher utilization, according to t
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
The US market narrative undergoes a "dramatic reversal": shifting from "AI-driven deflation and controllable US debt" to "AI squeezing the bond market, and Bassen unable to control long-term interest rates"
Deutsche Bank points out that the market narrative regarding the U.S. economy has shifted from "AI drives down inflation" to "AI-related bond issuance pushes up U.S. Treasury yields," but pessimism may be overdone. The bank believes that the real risks underestimated by the market are security incidents in the AI ecosystem, failed IPOs, or underwhelming revenues, which could undermine the dollar and support the bond market. In addition, France's fiscal difficulties are putting new pressure on the euro.

Deutsche Bank: The "fifth wave of tech stock rally" in US stocks since late July has peaked, prepare for a "V-shaped reversal"
Deutsche Bank has downgraded its rating on US technology stocks from overweight to neutral, noting that the fifth round of tech stock rally since July 29 is approaching the upper boundary of the long-term trend channel. The current upside potential is only about 4 percentage points, while historical trends indicate downside risks could reach 16 percentage points. Funds are expected to rotate into other sectors, and the European market, with its lower tech exposure, is likely to benefit relatively. However, Deutsche Bank emphasized that the long-term outperformance trend of technology stocks remains unchanged.
How expensive is AI computing power rental in the US? "Spot price" is twice that of long-term contracts, and four times the return threshold for cloud service providers.
The short-term spot leasing price for AI computing power reaches as high as $40 to $50 billion per gigawatt per year, while the price for long-term contracts is only $20 billion per year, and the breakeven threshold for supercomputing cloud operators is around $12 billion per year. However, according to Goldman Sachs, the fundamental reason why hyperscale cloud providers like Google rent computing power at such significant premiums is that their in-house capacity cannot keep up; once their own capacity catches up, the spot premium will disappear.
