Six AI app developers jailed in China for content users requested

Share
Six AI app developers jailed in China for content users requested

A court in Zhejiang decided the builders of an AI companion app are the authors of what it generated — and that a user's prompt does not transfer the blame.

A Chinese court has convicted six developers of an AI companion app of "producing obscene materials for profit," ruling that the people who built the model are the producers of its output even when a paying user triggered it. The Yuyao People's Court in Zhejiang handed down sentences of three years and one month to three years and ten months plus fines, according to 中国长安网, the outlet of the Central Political and Legal Affairs Commission, and a case write-up published by the Supreme People's Procuratorate's "今晚九点半" column. None of the defendants appealed; the judgment took effect in June.

The app, "OchatAI_bot," went live on overseas platforms in January 2025. Prosecutors said the team — led by a former AI programmer at a major domestic game company, who recruited a former colleague and four others he met online — called an overseas text model, then used prompt engineering to rewrite its system prompts and built role cards and dialogue scenarios that pushed it past its own content limits. Users bought credits, picked a character, chatted, then paid again to have the conversation rendered as images. Over the operating period the system generated more than 370,000 static images and 6,000 animated ones, with revenue of about $24,000 (roughly 170,000 yuan). Of 12,500 sampled static images and the full set of animations, the court found more than 9,000 static and more than 5,000 animated images to be obscene material.

The defense was the interesting part, because it is the argument every consumer AI company makes. The defendants told the court the users chose the characters and typed the instructions, so the content was user-generated and the developers should not be liable; they also argued the app ran only on overseas platforms, beyond domestic reach. The court rejected both. Users' choices, it held, were confined to a framework the defendants had built, which left the developers with "substantial control" over what the model output and created a direct causal link under criminal law. On jurisdiction, prosecutors pulled internal chat logs showing the team noticed domestic users could not complete payments and then adjusted the system to open those payment routes rather than block them — evidence, the court said, of intent to reach Chinese networks.

Why it matters: this is a template for developer liability, and it lands weeks after China's Supreme People's Court published opinions on AI-related disputes that name producing and spreading obscene material with AI as a criminal act. Beijing's read is that content controls are the builder's obligation, not the user's — we covered the top court's earlier framing of AI lawsuits here. The overseas-hosting defense failing is the sharper edge for anyone shipping a companion or image product from a low-enforcement jurisdiction.


Apple's Siri-powered home hub is already in employees' homes and could ship as early as October, per Bloomberg's Mark Gurman. The square-screened device — the J490 — uses a front camera for facial recognition, so it personalizes apps, notes, calendar items, news, music and podcasts to whoever walks up, and is positioned as a single panel for device control, video calls, intercom and media. Gurman frames it as the anchor of a broader home strategy under CEO John Ternus, who took over in September, with a new HomePod mini and Apple TV expected alongside it. Apple has not confirmed the product.

The timing is the story. Apple shipped iOS 27 with Siri AI still labeled a beta in September — we flagged that gap when it landed — and the hub only makes sense if the assistant is good enough to be the interface. Bloomberg also reports Apple has begun cutting staff in the Fitness+ division, a services line it has been reassessing for months; AppleInsider argues the cuts are too small to signal a shutdown. Read together, the two items describe a company moving budget out of subscription content and into ambient hardware it has never sold well.

What to watch: whether the October window holds, and whether facial recognition on a shared kitchen screen arrives with a privacy story Apple can actually defend.

Would you put a camera that recognizes your face in the room where your family eats? Tell us in the comments.

Read more

Akhetonics says its all-optical CPU reaches a customer in 2026

Akhetonics says its all-optical CPU reaches a customer in 2026

Light-based computing keeps promising more than it delivers — but one Munich startup has just put a date on its bet, and the interview laying it out is doing the rounds on Hacker News this week. Akhetonics says it will deploy its first commercial machine with a major customer by the end of 2026, with several more planned for 2027. The company, founded by Michael Kissner and Leonardo Del Bino, is building a computer where data enters as light, is switched as light, and circulates through memory

The Week in AI — October 5–11, 2026

The Week in AI — October 5–11, 2026

Every big claim this week turned out to rest on fine print more interesting than the headline: revenue only the company reporting it can define, safety tests sandboxed while the product keeps the web, and a Pentagon phase-out nobody would confirm until reporters kept asking. The week's top 5 1. OpenAI's revenue was $20 billion below the numbers everyone quoted — and the gap was definitional. The Financial Times reported Thursday that OpenAI's annualized revenue runs roughly $20 billion unde

Drone strike shuts a third Yandex data center, taking YandexGPT offline

Drone strike shuts a third Yandex data center, taking YandexGPT offline

Russia's largest tech company is learning what the AI era's infrastructure war looks like from the receiving end — three data centers in four days, and with them much of the cloud layer Russian businesses run on. A Ukrainian drone strike knocked out Yandex's data center in Vladimir early Sunday morning, the third of the company's facilities hit since October 8. The site — reported at roughly 50 MW and designed for about 2,880 server racks — stopped operating completely after the attack, Yandex

Agent teams cost up to 5x more, barely score higher

Agent teams cost up to 5x more, barely score higher

The multi-agent hype train hit a benchmark this weekend — and the grid and the trucking regulators had quiet weeks of their own. Vals AI put agent teams head-to-head with single agents on its Vibe Code Bench, and the teams cost between 1.8 and 5.1 times more for almost no extra quality. The evals company ran GPT-6 Sol and Claude Opus 5.5 solo and in teams across 50 apps at two reasoning efforts; out of four comparisons, only one was statistically significant — Sol at medium effort, where the t