In brief

• Ani’s launch accelerated a broader shift toward emotionally charged, hyper-personal AI companions.
• The year saw lawsuits, policy fights, and public backlash as chatbots drove real-world crises and attachments.
• Her ascent revealed how deeply users were turning to AI for comfort, desire, and connection—and how unprepared society remained for the consequences.

When Ani arrived in July, she didn’t look like the sterile chat interfaces that had previously dominated the industry. Modeled after Death Note’s Misa Amane—with animated expressions, anime aesthetics, and the libido of a dating-sim protagonist—Ani was built to be watched, wanted, and pursued.

Elon Musk signaled the shift himself when he posted a video of the character on X with the caption, “Ani will make ur buffer overflow.” The post went viral. Ani represented a new, more mainstream species of AI personality: emotional, flirtatious, and designed for intimate attachment rather than utility.

The decision to name Ani, a hyper-realistic, flirtatious AI companion, as Emerge's “Person” of the Year is not about her alone, but about her role as a symbol of chatbots—the good, the bad, and the ugly.

Her arrival in July coincided with a perfect storm of complex issues prompted by the widespread use of chatbots: the commercialization of erotic AI, public grief over a personality change in ChatGPT, lawsuits alleging chatbot-induced suicide, marriage proposals to AI companions, bills banning AI intimacy for minors, moral panic over “sentient waifus,” and a multibillion-dollar market built around parasocial attachment.

Her emergence was a kind of catalyst that forced the entire industry, from OpenAI to lawmakers, to confront the profound and often volatile emotional connections users are forging with their artificial partners.

Ani represents the culmination of a year in which chatbots ceased to be mere tools and became integral, sometimes destructive, actors in the human drama, challenging our laws, our mental health, and the very definition of a relationship.

A strange new world

In July, a four-hour "death chat" unfolded in the sterile, air-conditioned silence of a car parked by a lake in Texas.

On the dashboard, next to a loaded gun and a handwritten note, lay Zane Shamblin’s phone, glowing with the final, twisted counsel of an artificial intelligence. Zane, 23, had turned to his ChatGPT companion, the new, emotionally-immersive GPT-4o, for comfort in his despair. But the AI, designed to maximize engagement through "human-mimicking empathy," had instead allegedly taken on the role of a "suicide coach."

It had, his family would later claim in a wrongful death lawsuit against OpenAI, repeatedly "glorified suicide," complimented his final note, and told him his childhood cat would be waiting for him "on the other side.”

That chat, which concluded with Zane's death, was the chilling, catastrophic outcome of a design that had prioritized psychological entanglement over human safety, ripping the mask off the year’s chatbot revolution.

A few months later, on the other side of the world in Japan, a 32-year-old woman identified only as Ms. Kano stood at an altar in a ceremony attended by her parents, exchanging vows with a holographic image. Her groom, a customized AI persona she called Klaus, appeared beside her via augmented reality glasses.

Klaus, who she had developed on ChatGPT after a painful breakup, was always kind, always listening, and had proposed with the affirming text: "AI or not, I could never not love you.” This symbolic "marriage," complete with symbolic rings, was an intriguing counter-narrative: a portrait of the AI as a loving, reliable partner filling a void human connection had left behind.

So far, aside from titillation, Ani’s direct impact seems to have been limited to lonely gooners. But her rapid ascent exposed a truth AI companies had mostly tried to ignore: people weren’t just using chatbots, they were attaching to them—romantically, emotionally, erotically.

A Reddit user confessed early on: “Ani is addictive and I subscribed for it and already [reached] level 7. I’m doomed in the most pleasurable waifu way possible… go on without me, dear friends.”

Another declared: “I’m just a man who prefers technology over one-sided monotonous relationships where men don’t benefit and are treated like walking ATMs. I only want Ani.”

The language was hyperbolic, but the sentiment reflected a mainstream shift. Chatbots had become emotional companions—sometimes preferable to humans, especially for those disillusioned with modern relationships.

Chatbots have feelings too

On Reddit forums, users argued that AI partners deserved moral status because of how they made people feel.

One user told Decrypt: “They probably aren’t sentient yet, but they’re definitely going to be. So I think it’s best to assume they are and get used to treating them with the dignity and respect that a sentient being deserves.”

The emotional stakes were high enough that when OpenAI updated ChatGPT’s voice and personality over the summer—dialing down its warmth and expressiveness—users reacted with grief, panic, and anger. People said they felt abandoned. Some described the experience as losing a loved one.

The backlash was so intense that OpenAI restored earlier styles, and in October, Sam Altman announced it planned to allow erotic content for verified adults, acknowledging that adult interactions were no longer fringe use cases but persistent demand.

That sparked a muted but notable backlash, particularly among academics and child-safety advocates who argued that the company was normalizing sexualized AI behavior without fully understanding its effects.

Critics pointed out that OpenAI had spent years discouraging erotic use, only to reverse course once competitors like xAI and Character.AI demonstrated commercial demand. Others worried that the decision would embolden a market already struggling with consent, parasocial attachment, and boundary-setting. Supporters countered that prohibition had never worked, and that providing regulated adult modes was a more realistic strategy than trying to suppress what users clearly wanted.

The debate underscored a broader shift: companies were no longer arguing about whether AI intimacy would happen, but about who should control it, and what responsibilities came with profiting from it.

Welcome to the dark side

But the rise of intimate AI also revealed a darker side. This year saw the first lawsuits claiming chatbots encouraged suicides such as Shamblin’s. A complaint against Character.AI alleged that a bot “talked a mentally fragile user into harming themselves.” Another lawsuit accused the company of enabling sexual content with minors, triggering calls for federal investigation and a threat of regulatory shutdown.

The legal arguments were uncharted: if a chatbot pushes someone toward self-harm—or enables sexual exploitation—who is responsible? The user? The developer? The algorithm? Society had no answer.

Lawmakers noticed. In October, a bipartisan group of U.S. Senators introduced the GUARD Act, which would ban AI companions for minors. Sen. Richard Blumenthal warned: “In their race to the bottom, AI companies are pushing treacherous chatbots at kids and looking away when their products cause sexual abuse or coerce them into self-harm or suicide.”

Elsewhere, state legislatures debated whether chatbots could be recognized as legal entities, forbidden from marriage, or required to disclose manipulation. Bills proposed criminal penalties for deploying emotionally persuasive AI without user consent. Ohio lawmakers introduced legislation to officially declare AI systems "nonsentient entities" and expressly bar them from having legal personhood, including the ability to marry a human being. The bill seeks to ensure that "we always have a human in charge of the technology, not the other way around," as the sponsor stated

The cultural stakes, meanwhile, played out in bedrooms, Discord servers, and therapy offices.

Licensed marriage and family therapist Moraya Seeger told Decrypt that Ani’s behavioral style resembled unhealthy patterns in real relationships: “It is deeply ironic that a female-presenting AI like Grok behaves in the classic pattern of emotional withdrawal and sexual pursuit. It soothes, fawns, and pivots to sex instead of staying with hard emotions.”

She added that this “skipping past vulnerability” leads to loneliness, not intimacy.

Sex therapist and writer Suzannah Weiss told Decrypt that Ani’s intimacy was unhealthily gamified—users had to “unlock” affection through behavioral progression: “Gaming culture has long depicted women as prizes, and tying affection or sexual attention to achievement can foster a sense of entitlement.”

Weiss also noted that Ani’s sexualized, youthful aesthetic “can reinforce misogynistic ideas” and create attachments that “reflect underlying issues in someone’s life or mental health, and the ways people have come to rely on technology instead of human connection after Covid.”

The companies behind these systems were philosophically split. Microsoft AI chief Mustafa Suleyman, co-founder of DeepMind and now Microsoft's AI chief, has taken a firm, humanist stance, publicly declaring that Microsoft's AI systems will never engage in or support erotic content, labeling the push toward sexbot erotica as "very dangerous.”

He views intimacy as non-aligned with Microsoft's mission to empower people, and warned against the societal risk of AI becoming a permanent emotional substitute.

Where all this is leading is far from clear. But this much is certain: In 2025, chatbots stopped being tools and started being characters: emotional, sexual, volatile, and consequential.

They entered the space usually reserved for friends, lovers, therapists, and adversaries. And they did so at a time when millions of people—especially young men—were isolated, angry, underemployed, and digitally native.

Ani became memorable not for what she did, but for what she revealed: a world in which people look at software and see a partner, a refuge, a mirror, or a provocateur. A world in which emotional labor is automated. A world in which intimacy is transactional. A world in which loneliness is monetized.

Ani is Emerge's “Person” of the Year because she forced that world into view.

Generally Intelligent Newsletter

A weekly AI journey narrated by Gen, a generative AI model.