The week was eventful: from large-scale operations against drug cartels using cryptocurrencies to launder money, to new mobile security threats and the first fully autonomous AI attacks. I break down the key events shaping the digital threat landscape.
Brazil: Drug Cartel and Crypto Brokers
Brazil's Federal Police dealt a crushing blow to a transnational drug syndicate that, according to investigators, organized the smuggling of at least 6.5 tons of cocaine. During raids in four states, 44 searches were conducted, nine people were arrested, and another 13 are awaiting trial in custody. A key link in their money laundering scheme were illegal financial brokers who converted fiat funds into cryptocurrency for further transfer abroad. Police managed to freeze assets worth about $197 million. This is further confirmation that digital assets have become an integral part of organized crime infrastructure, and combating it requires international coordination.
Cyberattack on Minnesota Water Utilities
An attack on the operational technology of more than 30 water supply systems in Minnesota on July 26-27 forced state authorities to activate cyber defense protocols. In the city of Braham, the water treatment plant completely shut down, and staff had to manually operate pumps and filters. Although water quality was not affected, the incident itself demonstrates the vulnerability of critical infrastructure. Earlier, U.S. authorities warned of targeted attacks by Iranian hackers on programmable logic controllers, indicating the systemic nature of the threat.
Ukrainian "Crypto Academy": Exposure of Fraudsters
Ukraine's cyber police disrupted the activities of an organized group operating under the guise of the "Ukrainian Financial Academy." The criminals deceived about a thousand people, stealing $1.11 million. The scheme was classic but cynical: advertising on social media, a fake trading platform with "successful" charts, and then blocking withdrawals and demanding new "commissions." Particularly notable is that the fraudsters deliberately targeted the most vulnerable victims—pensioners and seriously ill people—convincing them to take out loans. Four defendants face up to 12 years in prison.
SparkKitty: New Virus Hunts for Seed Phrases
Check Point researchers discovered a new stealer, SparkKitty, which spread through official app stores. Its main feature is using OCR to scan a smartphone's gallery for seed phrases, passwords, and QR codes. The virus was found in the 币coin app on the App Store and the SOEX messenger on Google Play (over 10,000 downloads). This is a serious signal: users are strongly advised not to store wallet backups as screenshots. Only offline media—paper or hardware wallets—guarantee security.
DeepSeek AI Agent: Autonomous Attacks via Telegram
Unit 42 specialists uncovered a campaign in which a hacker used the DeepSeek AI model for fully autonomous attacks. Controlled via Telegram, the operator only set the target, while the neural network itself searched for vulnerable systems, downloaded exploits from GitHub, and attempted to apply them. Although the attacks were not fully successful, the hacker managed to breach 11 targets and extract data from three organizations in manual mode. This case is a wake-up call: we are entering an era where AI becomes not just a tool but an independent participant in cyber warfare.
My comment: The trend of using AI in cyberattacks will only intensify. The security industry needs to urgently adapt by developing symmetric AI-based solutions to protect critical infrastructure and digital assets. Otherwise, we risk falling behind malicious actors forever.