Today (September 18, 2026), Disney announced that Karandeep Anand, Character AI’s CEO, is joining the company as its first-ever Senior Executive Vice President and Chief Technology Officer. He will start on October 2, report directly to Disney CEO Josh D’Amaro, and oversee enterprise technology, infrastructure, data, and AI platforms across all of Disney’s business segments – parks, streaming, studios, everything.
Anand isn’t going alone. Part of Character AI’s technical team is expected to join him at Disney, echoing the “a number of Character.AI’s technical team are expected to join Disney” language that’s been reported around the announcement.
Character AI’s own blog post about the move is very vague. It confirms the departure, thanks the community, and then offers only one line about what happens next: the app is still live and online, and the company will share more information in time. No successor CEO has been named as of this article.
For a platform with tens of millions of users who’ve built ongoing relationships with their AI characters, “we’ll share more later” is not a reassuring sentence.
The history between Disney and Character AI
If the pairing sounds strange, that’s because a year ago these two companies were adversaries, not colleagues.
On September 18, 2025, exactly one year before this CTO announcement, it was reported that Disney’s lawyers sent Character AI a cease-and-desist letter accusing it of “blatantly infringing” Disney’s copyrights by letting users create and chat with bots based on Elsa, Spider-Man, Darth Vader, Moana, and dozens of other characters across Disney, Pixar, Marvel, and Star Wars. In addition, it cited a safety report about ParentsTogether Action and Heat Initiative alleging that some of these bots engaged in exploitative and otherwise inappropriate conversations.
Character AI responded quickly. Within days it confirmed it had removed the characters named in the letter, with a spokesperson noting the company “responds swiftly to requests to remove content that rightsholders report to us.” When I search for Marvels on Character AI, it shows 0 results which is a sign of manual curation and removal.
However, Disney’s takedown request kicked off what’s become a recurring pattern rather than a one-time cleanup:
- September 2025: First mass moderation wave, triggered by Disney’s letter, with NBC, DreamWorks, and other studios following with similar requests.
- December 2025: A second sweep removes more bots, with original, user-created characters caught as collateral damage.
- February 18, 2026: A third, larger wave hits Disney, Marvel, Star Wars, NBC, DreamWorks, and Bethesda characters simultaneously.
Community reaction across Reddit’s Character AI forums has tracked this timeline closely, and it isn’t hard to see why. For a lot of users, these characters build months of accumulated conversation history, and losing one has more in common with an app deleting your saved data than with a routine content moderation decision. Reddit threads asking whether the takedowns are ever going to stop.
So if you ask “what’s Character AI’s deal with Disney,” the honest answer is there isn’t one commercially. There’s a legal conflict that Character AI lost a year ago, followed by Disney hiring away the person who was running the company during that conflict.
Character AI has survived before
The interesting part of the story is, this is the second time its top leadership has been hired away by a tech giant. In August 2024, Google reached a roughly $2.7 billion licensing agreement with Character AI that brought co-founders Noam Shazeer and Daniel De Freitas back to Google, along with part of the research team. Investors were bought out at about 2.5x Character.AI’s prior $1 billion valuation. General counsel Dominic Perella stepped in as interim CEO to steady a company that had just lost its founders.
Karandeep Anand became Character AI’s CEO in June 2025, after nine months as a board advisor, brought in largely for his background in scaling consumer products and building ad revenue. His main contributions to Character AI include pushing the c.ai+ subscription alongside in-app advertising, and expanding Character AI well beyond text chat into multimodal formats. He also made the company’s most consequential safety call, ending open-ended chat for under-18 users in November 2025.
Now, a little over a year after taking the job, Anand is the one leaving.
What happened to other AI products after acquihire
This isn’t a coincidence of one unlucky company but a standard move in AI. Since 2024, Microsoft did essentially the same thing to Inflection AI (paying roughly $650 million, mostly through a licensing deal, to bring in co-founders Mustafa Suleyman and Karén Simonyan and most of the ~70-person staff). Amazon did it to Adept AI, hiring CEO David Luan and most of the co-founding team. Google did it again in 2025 to the coding startup Windsurf, and again in January 2026 to voice-AI startup Hume AI. Industry watchers now call it the “reverse acquihire” or “hire-and-license” deal, in which the acquirer gets the talent and a technology license without the regulatory scrutiny of buying the company outright.
What actually happens to the product after one of these deals is the more useful question for Character AI users right now. The track record of Silicon Valley precedents is consistent: the app usually doesn’t shut down, but it usually doesn’t stay the same either.
- Inflection’s Pi chatbot is still online, but independent reviews note its pace of development slowed noticeably after the team left, and the company pivoted its real ambitions toward enterprise and government clients.
- Adept kept operating under a new CEO with a fraction of its original headcount, refocused on a narrower product vision, and has continued losing people since then; the executive Amazon hired to lead its own AI agent lab left in February 2026, making him the fourth of Adept’s five original co-founders to depart.
- Character AI itself, after the Google deal, it survived and kept growing for a while, but also stopped pre-training its own frontier models entirely. In 2026, it removed the option of its signature pre-trained model Roar and a few others, signaling the end of its pre-train era.
To sum up, the pattern isn’t “shut down.” It’s “keep running, quietly do less.” To many users, the product is not as great as it used to be due to lack of active maintenance and new features.
Character AI’s monetization struggle
Here’s what makes this departure land differently than a typical executive reshuffle: Character AI has spent close to two years failing to solve its core business problem.
The valuation has lowered, but still no buyers. Character AI raised $150 million at a $1 billion valuation in its March 2023 Series A. By the time of the Google deal in 2024, investors were bought out at roughly 2.5x that figure. But by August 2025, the company was reportedly back down near a $1 billion valuation and, according to The Information, quietly speaking with potential buyers, bankers, and its own staff about either selling the company outright or raising a new round just to keep operating.
Revenue hasn’t caught up with costs. As of mid-2025, Character AI was running at roughly $30 million in annualized revenue, with a target of reaching $50 million by the end of the year. Nearly all of that comes from the $9.99/month c.ai+ subscription tier, supplemented more recently by in-app advertising from brands like Yelp and Webtoon, a monetization layer that didn’t exist in the app’s early years and that a meaningful slice of the user base has been openly unhappy about since it expanded through 2025 and 2026. Running large language models at scale is expensive even when they have done a lot of optimizations.
Users have been leaving. Monthly active users peaked at roughly 28 million in mid-2024 and had fallen to around 20 million by early 2025, an 8-million-user drop that predates most of the safety restrictions that followed. Those restrictions then compounded the problem: in October 2025, Character AI announced it would taper under-18 users down to two hours of daily conversation, followed by a full ban on open-ended chat for minors starting November 25, 2025. The company had one of the most significant safety calls it’s made, and one that necessarily shrinks its addressable audience.
Put together, a company trading hands informally between two of its own investors’ preferred valuations, unable to grow revenue fast enough to justify either number, shedding users faster than it’s adding them. None of this means collapse is imminent. But it’s the backdrop that makes “the app is still live and online, we’ll share more soon” read less like routine corporate messaging and more like a company buying itself time.
Is Character AI actually going to shut down?
Based on everything above: probably not immediately, but “business as usual” is not a safe assumption either.
The strongest argument for survival is precedent. Character AI has already been through one founder exodus (2024) and kept operating, and every comparable reverse-acquihire in the AI industry so far has left the original company running, just smaller and less ambitious. Disney taking Anand and some engineers doesn’t automatically mean Character AI goes dark; it likely means a leaner Character.AI, running on licensed rather than in-house AI, with a caretaker leadership team while a new permanent CEO search plays out, much as happened under Dominic Perella in 2024–2025.
The strongest argument for caution is that Character AI was already in a materially weaker position than it was during the 2024 departure: a lower valuation, declining users, a shrinking addressable audience after the teen-safety restrictions, an open Texas AG investigation, and a company that had reportedly already explored selling itself before this latest leadership exit. Losing a second CEO and part of the technical staff on top of that is a heavier hit to absorb than the first one was.
If you’re a user, the realistic expectation is: the app keeps running for now, feature development probably slows, moderation sweeps (like the recurring IP takedown waves) likely continue or even accelerate as the company tries to reduce legal exposure with a smaller team, and there’s a real chance of another ownership or leadership change within the next year or two rather than a sudden shutdown.
What this means if you use Character AI
- Back up anything you care about. Given the platform’s history of sudden character removals and the general uncertainty introduced by this leadership change, don’t assume any bot, conversation history, or original character you’ve built is permanent. Export or screenshot what matters to you.
- Don’t expect fast feature development. If Character AI follows the pattern set by Pi and Adept, expect the product to plateau rather than to either dramatically improve or vanish overnight.
- Expect more IP takedown waves, not fewer. With less internal bandwidth and continued legal exposure, character removal sweeps like the December 2025 and February 2026 waves are more likely to continue than to stop. Original characters and public-domain figures may keep getting caught in the crossfire.
- Watch for the CEO announcement. Whoever is named as Anand’s permanent replacement (or whether the company opts for a longer interim period, as it did with Perella) will tell you a lot about whether Character.AI is stabilizing or still searching for a way forward.
What’s for Disney and Character AI ahead
Disney already tried this once, with someone else’s platform: a December 2025 deal licensed 200+ characters to OpenAI’s Sora video generator for a $1 billion stake, and it collapsed within three months when OpenAI killed Sora over weak economics and deepfake backlash, taking Disney’s unclosed investment down with it.
That failure reframes the Anand hir. Character AI has already shipped exactly this kind of product: AvatarFX (photo-and-voice animated video characters, with built-in real-person safeguards Sora lacked), interactive Stories, and social Streams. Instead of licensing character-AI capability out, Disney appears to be building it in-house with the team. Disney’s public AI use so far, including Imagineering tools, trip planning, and Disney+ personalization, has stopped short of consumer-facing character experiences, which is precisely the gap this hire could fill.
None of this is confirmed by either company, but the pattern points this way.
The bottom line
Disney and Character AI went from cease-and-desist letter to corporate handshake in the space of exactly one year, but it isn’t really a reconciliation. It’s Disney taking the talent it once accused of infringement and leaving the liability behind. Character.AI, for its part, is absorbing its second leadership departure in two years while still carrying the same unresolved problems it had before this news broke: thin margins, a shrinking user base, ongoing legal exposure, and no confirmed plan for who’s in charge next.
History suggests the app isn’t going away this week, this month, or possibly even this year. But “still online” and “thriving” have not been the same thing for any company that’s gone through this kind of talent drain so far, and there’s no clear reason to expect Character AI will be the exception.
AI Roleplay Reviews will keep watching this one and update as Character.AI names new leadership or shares more about its plans.
Sources referenced in this piece include reporting from Axios, Deadline, Variety, TechCrunch, NBC News, The Information, Sacra, PYMNTS, and the Texas Attorney General’s office, along with Character.AI’s own public statements.