News
Password & Crypto-Stealing Trojan Targets UAE Users Via App Stores
The newly-discovered malware, named SparkCat, uses optical recognition to scan for sensitive data inside screenshots and photos.
Researchers at Kaspersky have identified a Trojan lurking in both the App Store and Google Play since at least March 2024. The malware is known as SparkCat, and it uses optical recognition to scan image galleries, stealing sensitive data like crypto wallet recovery phrases and passwords from user’s screenshots and photos.
How SparkCat Spreads
The malware is being distributed through both infected legitimate apps and cleverly disguised lures. These include messaging apps, AI assistants, food delivery services, and crypto-related applications. Some of the compromised apps have made their way into Google Play and the App Store, while others are circulating through unofficial sources.
Who’s At Risk?
Worryingly for our readers, the primary targets appear to be users in the United Arab Emirates. However, various countries in Europe and Asia have also been targeted, as SparkCat’s optical scanning can detect multiple languages.
How It Works
Once installed, SparkCat may request permission to access a user’s photo gallery. From there, it uses an optical character recognition (OCR) module to scan stored images for relevant keywords. If it detects sensitive data — particularly screenshots of recovery phrases for cryptocurrency wallets — it transmits the images to attackers. With this information, hackers can gain full access to a victim’s funds.
Who’s Behind It?
An analysis of the Android version of SparkCat revealed code comments written in Chinese, while the iOS variant included home directory names like qiongwu and quiwengjing. While these clues suggest a Chinese-speaking threat actor, there isn’t enough evidence to link the software to any known cybercriminal groups.
Sergey Puzan, a malware analyst at Kaspersky, noted, “This is the first known case of OCR-based Trojan to sneak into AppStore. In terms of both AppStore and Google Play, at the moment it’s unclear whether applications in these stores were compromised through a supply chain attack or through various other methods. Some apps, like food delivery services, appear legitimate, while others are clearly designed as lures”.
How To Protect Yourself
To minimize the risk of infection, Kaspersky recommends taking the following precautions:
- If you’ve installed an affected app, delete it immediately and avoid using it until a safe update is available.
- Avoid saving screenshots that contain sensitive information, such as crypto wallet recovery phrases. Instead, use secure password managers.
- Consider using reputable cybersecurity solutions like Kaspersky Premium to safeguard against malware threats.
As and mobile malware tactics evolve, staying cautious with app downloads and maintaining strong security practices can go a long way in keeping your data safe.
News
Nano Banana 2 Arrives In MENA For Google Gemini Users
Google brings its latest image model to Gemini and Search, adding 4K output and tighter text control for regional users.
Google has opened access to Nano Banana 2 across the Middle East and North Africa, pushing its newest image model into everyday tools rather than keeping it inside the exclusive (and expensive) Pro tier.
The rollout spans the Google Gemini desktop and mobile apps, and extends to Google Search through Lens and AI Mode. Developers can also test it in preview via AI Studio and the Gemini API.
Nano Banana 2 runs on Gemini Flash, Google’s fast inference layer. The focus is speed, but also control. Users can export visuals from 512px up to 4K, adjusting aspect ratios for everything from vertical social posts to widescreen displays.
The model maintains character likeness across up to five figures and preserves fidelity for as many as 14 objects within a single workflow. This enables visual continuity across scenes, iterations, or edits — supporting projects like short films, storyboards, and multi-scene narratives. Text rendering has also been improved, delivering legible typography in mockups and greeting cards, with built-in translation and localization directly within images.
Also Read: RØDE Adds Direct iPhone Pairing To Wireless GO And Pro Mics
Under the hood, the system taps Gemini’s broader knowledge base and pulls in real-time information and imagery from web search to render specific subjects more accurately. Lighting and fine detail have been upgraded, without slowing output.
By embedding the model inside Gemini and Search, Google is normalizing advanced image generation for a mass audience. In MENA, where startups and marketing teams are leaning heavily on AI to scale content across languages and borders, that shift lands at a practical moment.
The move also folds creative tooling deeper into search itself, so that image generation is no longer a separate workflow. It now sits right next to the query box.
