9to5Mac58%
Investigation reveals dozens of disguised gambling apps on the App Store in Brazil 48%
By 9to5 Staff0%
7/17/2026, 2:43:08 PM
BS Summary: This article contains 19 faulty reasoning types, including Biased Writer Voice, Recency Bias, and Post Hoc (False Cause), with Negativity Bias as the most egregious example at 15.1% saturation with 83 hits. Analysis detected 868 faulty-reasoning hits from 549 analyzed words, generating a BS Score of 49% and a BS Rank of 48% (11,516 of 21,887 articles). This article is better (less manipulative) than 52.60% of the article peer group.
An investigation by 9to5Mac reveals dozens of apps that disguise gambling platforms as simple games and utilities.
Here are the details.
Betting platforms pose as simple apps
Brazilian users browsing the App Store rankings in categories such as Navigation, Travel, and Weather have noticed a growing number of poorly made games appearing among the top results, many of them featuring AI-generated illustrations of animals as their app icons.
As it turns out, these are so-called jacket apps, which are just a front for hidden betting and gambling apps.
A 9to5Mac investigation has uncovered more than 60 apps that behave exactly as depicted in their App Store screenshots when accessed from virtually anywhere in the world, except Brazil.
When opened from a Brazilian IP address, the same apps instead reveal online betting platforms, as shown in the example below, which is currently the top app in the Weather category:
Most of the apps are published by developer accounts with only a single App Store listing.
Many of the developer names appear to be common in Vietnam and other countries, rather than Brazil.
The apps also tend to use similar (and, in some cases, identical) privacy policies, have generally no recorded updates, and are roughly 15MB in size.
Digging deeper, 9to5Mac found a public GitHub repository containing instructions for a Cursor agent to create simple, vibe-coded apps that serve as fronts for the betting platforms.
The instructions call for each app to include three to five visible interfaces, use a marketable name and animal-themed icon (a dragon, ox, rabbit, rat, or tiger), and support remotely controlled routing to either the local app, an in-app web page, or an external website.
They also say the apps should be built around simple, immediately understandable concepts, with clear branding and several prominent feature areas, while differing enough from one another to appear as separate products.
The repository also includes explicit instructions intended to prevent the apps from being flagged as suspicious during App Store review.
These include giving each app uniquely named startup and remote-configuration codes, making it harder for reviewers to identify them as part of the same group.
Ironically, the App Store’s own recommendation system appears to have little trouble grouping the apps together, as the “You Might Also Like” section on several of these apps points users toward other suspicious apps more frequently than they point to genuine apps.
Although this is far from a new problem on the App Store, the discovery comes on the heels of renewed pressure from Brazilian authorities over the availability of unauthorized betting apps on Apple’s and Google’s platforms.
Just a few days ago, Brazil’s Ministry of Justice gave both companies five business days to explain how they detect apps that hide or change betting features after approval, verify that operators are federally authorized, and prevent minors from accessing gambling services.
Earlier today, Apple was also ordered to remove eight AI “nudify” apps from the App Store after pressure from the San Francisco City Attorney, which came six months after a separate Tech Transparency Project investigation uncovered dozens of similar apps on the platform.
9to5Mac has reached out to Apple for comment, and we will update this story if we hear back.
Speakers
2speakers24%attributed speech416writer words
Voice mapSelect a segment to jump to its words
Selected voice
100%flagged-word coverage9to5Mac
91 attributed words68% of attributed speech77% writer coverage
Attribution is sentence-level. Pattern percentages are calculated only from words assigned to that voice.
Loading…
Loading…
Loading…
Analysis
Hover over highlighted words in the article to view the associated bias or fallacy analysis.