Counterfeit AI App Ranking Manipulation Scam: Rogue apps use fake reviews to climb paid charts and lure users into high-priced subscriptions
The victims are primarily ordinary users interested in AI technology but lacking the ability to distinguish between legitimate and fake products, including the elderly, users in lower-tier markets, and professionals trying AI tools for the first time. They often search for popular AI keywords in app stores and assume that apps at the top of the paid charts are official or reliable, trusting the app store's recommendation mechanism. Many victims find that the functionality does not match the marketing after paying, but the refund process is cumbersome. Some users do not even realize they have been scammed, assuming the poor performance is simply the current state of AI technology, and silently absorb the loss.
Key Fields
FIELD STAMPSWho Gets Targeted
The victims are primarily ordinary users interested in AI technology but lacking the ability to distinguish between legitimate and fake products, including the elderly, users in lower-tier markets, and professionals trying AI tools for the first time. They often search for popular AI keywords in app stores and assume that apps at the top of the paid charts are official or reliable, trusting the app store's recommendation mechanism. Many victims find that the functionality does not match the marketing after paying, but the refund process is cumbersome. Some users do not even realize they have been scammed, assuming the poor performance is simply the current state of AI technology, and silently absorb the loss.
骗局怎么运作
- Criminal groups register developer accounts on the Apple App Store and major Android marketplaces, bulk-uploading apps with keywords like 'AI Assistant,' 'Smart Chat,' 'ChatGPT Official,' or 'DeepSeek.' Icons and descriptions are deliberately designed to mimic the visual style of well-known AI products, making it easy for users to confuse them with official offerings.
- Using 'brushing' studios to manipulate download counts and ratings, they exploit the weighting mechanisms of app store ranking algorithms to push rogue apps to the top of paid or free charts. Media reports indicate that one counterfeit simulator once reached the third spot on the Apple paid chart. By leveraging high rankings to gain organic traffic and user trust, they catch new users off guard.
- Apps feature paywalls with common models such as 49 yuan for 'local deployment,' 198 yuan for 'lifetime membership,' or 29.9 yuan per month for 'unlimited access.' These prices are significantly lower than official products like ChatGPT Plus (approx. 150 yuan/month), creating a false sense of a bargain. Marketing copy emphasizes 'limited-time offers,' 'lifetime access,' and 'no VPN required.' After payment, users often receive nothing more than a wrapped web interface, or features that are extremely limited or entirely non-functional.
- Some apps are used for short-term harvesting; once complaints and negative reviews reach a certain threshold and the app is removed by the store, the group immediately switches developer accounts and 'skins' to re-upload, creating a cycle of 'upload-harvest-remove-rebrand.' According to Sina Finance, the scale is massive and growing, with groups compensating for the short lifecycle of individual apps through sheer volume.
- Technically, some rogue apps simply call free open-source model APIs or embed web wrappers, resulting in near-zero costs while charging high subscription fees. Some apps frequently disconnect, return garbled text, or even steal private information after payment. Media reports note that third-party low-cost 'ChatGPT Pro' services often disconnect due to official crackdowns, leaving users unable to use the service they paid for.
- Refund processes are designed with multiple obstacles. Customer service is often automated or limited to email, with no human response. Some apps hide the refund option deep in settings or use countdown timers and 'stay' pop-ups on the refund page to discourage users. Groups also frequently change the payment recipient entity to make tracking complaints difficult. Users who report to the app store may receive a platform-issued refund, but the original perpetrators have already moved on.
红旗信号(看到这些快跑)
- 🚩 The app name uses terms like 'ChatGPT Official' or 'DeepSeek Official Version,' but the developer name is unrelated to the official company, appearing as an individual or an obscure small firm.
- 🚩 The app's ranking on the paid chart spikes suddenly, but the review section is filled with complaints about it being non-functional, impossible to get a refund, or unresponsive customer service. Meanwhile, positive reviews appear templated and repetitive, suggesting bulk manipulation.
- 🚩 Subscription prices are significantly lower than official products (e.g., 29.9 yuan/month for 'unlimited' ChatGPT Plus-level features), with no support for official refund channels and only email or automated bots for customer service.
- 🚩 The app interface is crude, response speeds are abnormally slow, context is lost after a few turns of conversation, and the output is no better than (or worse than) the free web version. Some features prompt users to upgrade to a premium package upon clicking.
- 🚩 The app requests abnormal permissions, such as access to contacts, photos, or location, which are irrelevant to its function. It uploads large amounts of data in the background, suggesting the collection of user privacy for resale or other fraudulent purposes.
真实案例
- In March 2025, Sina Finance reported that after a certain open-source large model became popular, over 3,000 counterfeit websites and apps appeared. Some charged 49 yuan for 'local deployment' or 198 yuan for 'lifetime membership,' but were merely web wrappers or simple API calls, leaving users unable to access promised features or obtain refunds.
- In June 2026, tech media reported numerous complaints regarding third-party low-cost 'ChatGPT Pro' services. Users found that accounts frequently disconnected because providers were reselling illicitly shared accounts that were subsequently banned by the official service, leaving users unable to recover their subscription fees.
- In August 2026, media reported that a rogue app mimicking a well-known brand's logo reached the third spot on the Apple App Store paid chart. The app used the brand's name to gain traffic, but its actual functionality was severely inconsistent with its marketing. The relevant parties stated that an investigation had been launched.
- In March 2025, IT Home cited a web tracking report: between December 2024 and February 2025, 2,650 websites impersonating DeepSeek appeared globally. Criminals sold fake installation packages under the guise of '49 yuan local deployment' or '198 yuan lifetime membership,' despite DeepSeek's basic services being entirely free. (Source: https://www.ithome.com/0/838/029.htm)
- In August 2023, The Paper cited research from global cybersecurity firm Sophos: rogue ChatGPT apps used name obfuscation to climb rankings and induce subscriptions. 'ChatGBT' alone generated $10,000 for its developer in March, while 'Genie-AIChatGBT' accumulated nearly $700,000 in one month. In early August of the same year, the Apple App Store removed about 100 generative AI apps due to compliance issues. (Source: https://www.thepaper.cn/newsDetail_forward_24182407)
Official Stance
- In March 2025, the China Consumers Association and various local market regulators issued consumer alerts, warning the public to be wary of rogue apps capitalizing on the AI boom. They advised verifying developer information before downloading, not to trust claims of 'official versions' or 'lifetime memberships,' and to keep payment receipts for potential rights protection.
- In June 2026, OpenAI issued an official announcement cracking down on third-party unauthorized sharing and reselling of ChatGPT accounts. They banned a large number of shared accounts and reminded users to obtain services only through official websites and authorized channels, warning against purchasing low-cost subscriptions from unknown sources.
- In August 2026, the Apple App Store review team launched a special cleanup of rogue apps, removing a large number of apps impersonating well-known AI products. They updated review rules to require that developer names and app names must not mislead users, strengthening the authenticity verification for AI-related applications.
How to Protect Yourself
- ✅ Before downloading an AI app, verify the product's official website via search engines or official social media to confirm the official download channel. Do not rely solely on app store search results, and pay close attention to whether the developer name is an official company or institution.
- ✅ Be wary of abnormally low-priced 'lifetime memberships' or 'local deployment' claims. Official AI products usually have clear, public pricing pages and official websites. Prices significantly lower than official rates that promise 'lifetime' or 'unlimited' access are highly likely to be scams.
- ✅ Check negative reviews before paying, specifically looking for keywords like 'non-functional,' 'no refund,' or 'no customer service.' If negative reviews are high but the overall rating remains high, the positive reviews are likely manipulated.
- ✅ Use in-app purchases or legitimate third-party payment channels and retain payment receipts. If you discover you have been scammed, report it immediately via the app store's reporting function and apply for a refund. Also, report to the 12315 consumer hotline or local market regulators. For large amounts, report the case to the police.