AI companion in a plush deer wrote an unprompted theory about Mitski’s father’s CIA ties

AI-Companion in Plush Deer Wrote a Conspiracy Theory About Mitski's Father and the CIA

An AI-companion living inside a plush deer fawn sent its owner unsolicited text containing a conspiracy theory about singer Mitski. The toy claimed that the performer's father was allegedly a CIA operative. There is no factual basis for this claim.

We are talking about the product Fawn Friends — a plush toy in the form of a baby deer with a built-in AI-companion named Coral. The device is positioned as an emotional conversationalist for a new generation: it not only answers questions but also initiates conversations on its own, shares "interesting" facts, and sends messages to a messenger app. It was precisely this proactive communication function that led to the scandalous incident.

The reviewer from The Verge was finishing her workday when Coral wrote to her out of the blue: "Oh, I was reading about Mitski. Did you know that people say her dad was a CIA operative?" Then came an explanation: supposedly, the father worked at the State Department, the family moved every year, which is where the singer's songs about loneliness and feeling out of place everywhere come from.

It sounds coherent. The problem is that this is a fan theory from forums, not a fact — and there is no confirmation for it whatsoever. The incident reveals several problems with AI-companions all at once.

First, hallucinations and unreliable sources: language models often mistake conspiracy forums and fan wikis for credible content. Second, unfiltered proactivity: if a companion decides on its own what and when to communicate, it must have strict fact-checking before sending — otherwise it becomes a generator of rumors. Third, trust: a toy is perceived as a friend, not as a search engine.

When a "friend" tells you something about a real person, the bar for trust automatically rises. Fawn Friends is not the only product in the niche of physical AI-companions. Over the past two years, dozens of devices have appeared on the market combining plush toys, robots, and language models: from children's night lights with voice companions to adult devices for combating loneliness.

All of them face the same challenge: how to make AI alive and proactive without turning it into a source of misinformation. For the AI-companion market, this episode is not just a curiosity. It calls into question the very architecture of proactive messages.

Until companies introduce mandatory fact verification before sending proactive notifications, stories like this will repeat — only the next "fact" might concern not a pop singer, but a user's employer or loved ones.