Smart Engines and Intek Deliver Russian Scanners for AI-Powered Document Verification and Fraud Prevention
Smart Engines and GK Intek have introduced a family of Russian software-hardware scanner complexes designed for fast, secure document input and fraud detection. The devices handle passports, driver licences and other identity documents from Russia, CIS states and 235 additional jurisdictions.
The scanners perform automatic data capture and multi-spectral authenticity verification in a single pass. They examine documents under visible, ultraviolet and infrared illumination, analyse holograms and other security features, and read RFID chips when present. All processing occurs locally, eliminating the need to transmit images or personal data to external services.
Banks have already integrated the complexes to strengthen KYC and anti-fraud controls. At Genbank the solution reduced passport data entry to fractions of a second with 99.9 % accuracy. Severgazbank deployed the system to detect forged documents from Russia and 235 other countries, reporting a measurable drop in attempted fraud after integration in 2022.
At major airports the same technology operates inside the Sapsan border-control system. Modules equipped with the Sherlock neural ensemble now process passengers at Sheremetyevo, Vnukovo and Koltsovo, extracting data from Russian foreign passports, checking more than 600 authenticity features and identifying sophisticated fakes including deepfakes. After ten modules were installed at Sheremetyevo, average border-crossing time fell to 40 seconds and throughput increased six-fold.
Additional deployments cover automatic pass kiosks at seaports and river ports, electronic-signature issuance windows at Federal Tax Service offices, and the first pharmacy robot in Minsk that verifies customer age before dispensing restricted medicines. The scanners also support domestic hardware platforms including Komdiv, Elbrus and Baikal processors together with Russian operating systems.
Two U.S. patents further strengthen the offering: one detects hidden digital edits by analysing an image’s digital “handwriting” rather than its visual content; the second identifies optically variable devices from visible-light images taken under varying illumination angles, exposing copies and replicas.
Related articles
Silent Call Answering on Android: Defeating Phone Spam by Removing Human Attention
A detailed proposal suggests abandoning traditional spam call blocking in favor of allowing all incoming calls to connect automatically while keeping them invisible to the user. The approach uses Android's Telecom Framework and InCallService to answer calls silently without ringing, notifications, or screen activation. This breaks the economic model of mass dialing systems by inflating answered call metrics with empty connections that contain no human. The concept separates the technical establishment of a call from delivering user attention, forcing spammers to detect real people after the connection is made. Implementation requires the app to hold the ROLE_DIALER role and selectively invoke Call.answer() based on custom rules instead of always showing the incoming call UI. The author argues this shifts the detection burden onto robocall platforms and reduces the value of every successful connection.
Telegram Scam Bot Exposed by Fixed Timer and Deleted Messages in Telethon Userbot Analysis
A detailed investigation into a romance scam attempt on Telegram revealed an AI-driven userbot masquerading as a woman named Maria from Yaroslavl. The bot maintained consistent 4-5 minute response delays regardless of message length or time of day, responded to deleted messages, and accumulated multiple inputs before replying in batches. It refused out-of-character requests using repetitive phrases like "I am not a..." and handed off media or confusing inputs to a human operator. The bot failed to react to a nonexistent city name and ignored voice messages containing silence, leading to delayed human intervention. The chat was later deleted from the scammer side after testing, and the account ignored messages from a second profile. The analysis includes a full reconstruction of the bot's logic using the Telethon library, highlighting prompt protections against jailbreaks and reliance on fixed delays.
CACTER Upgrades PhishSim Anti-Phishing Simulation System to Reduce Employee Click Rates
CACTER has released an updated version of its PhishSim anti-phishing training platform that allows organizations to run realistic simulated attacks in just four steps. The system replicates common phishing vectors including malicious links, infected attachments, and disguised QR codes while spoofing sender addresses and official domains. Organizations can draw from a continuously refreshed template library covering invoices, financial subsidies, system notifications, and industry-specific scenarios. After each campaign the platform produces detailed visual reports that rank departments, classify employee risk levels, and recommend concrete remediation steps. Long-term use of the platform has been shown to lower average click rates from 23.88 percent to 4.16 percent. The solution is designed for immediate deployment without requiring dedicated security staff.
Scammers Abuse Custom GPT on ChatGPT.com to Deploy Windows RAT via ClickFix Technique
Researchers at Huntress uncovered a phishing campaign that leveraged a custom GPT named Plus 5.6 hosted directly on the official ChatGPT.com domain. Victims searching for ChatGPT were directed to the malicious GPT through sponsored Google results, where the bot instructed them to visit a backup domain due to alleged service issues. The link led to a Google Sites page mimicking a Cloudflare security check that triggered the ClickFix social engineering tactic. Users were prompted to copy and execute a command in Windows, initiating a multi-stage infection with a remote access trojan capable of full system control, file access, screen viewing, and camera or microphone activation. Huntress confirmed at least 40 incidents tied to the campaign, though only two infections were directly traced to the malicious GPT. The first GPT was removed on September 25 after notification, but a replacement linked to the same operation appeared by September 27.