WhatsApp Begins Limited Beta Testing of On-Device Scam Alert to Detect Fraud While Preserving End-to-End Encryption
WhatsApp, owned by Meta, has launched limited beta testing of its Scam Alert feature designed to warn users about potentially fraudulent messages from strangers.
The protection relies on a machine learning model that runs directly on the smartphone. Conversation content is never transmitted to WhatsApp or Meta, preserving end-to-end encryption. After activation, the device downloads the model and examines dialogue structure along with linguistic signs of fraud.
When a message appears suspicious, the recipient sees a warning while the sender remains unaware. Users can then block the contact, file a report, dismiss the alert, or designate the chat as trusted. In the trusted case, the last five messages can optionally be shared to help improve the model.
To prevent Meta from deploying a customized analyzer to specific users, every model release is recorded in an immutable log. A manifest containing SHA-256 hashes is signed with Ed25519 keys controlled by Cloudflare. The smartphone verifies the signature, log entry, and file integrity before executing the model.
WhatsApp collects only anonymized counters of detections and subsequent user actions. The app also includes a Scam Alert Activity log showing check results and model versions.
The current release remains an early technical preview. Meanwhile, Signal has introduced automatic key verification that uses a cryptographically verifiable log and allows users to confirm no unauthorized intermediary is present. Independent audits are performed by Cloudflare and Trail of Bits without access to user data.
Related articles
Russians Create Digital Cheat Sheets to Shield Relatives from Online Scams, Kaspersky Lab Study Reveals
A new study by Kaspersky Lab shows that 54% of Russians are actively concerned about their relatives' online safety and are taking concrete steps to protect them from fraud. The primary fear is scammers, cited by 58% of respondents, followed by financial losses at 48%. Instead of simple warnings, 63% now produce detailed digital or physical cheat sheets, with screenshots featuring arrows being the most popular format at 31%. Generation Z leads this trend, with 53% creating such guides compared to 46% of Millennials and 27% of Baby Boomers. The findings come from a 2026 survey of 1,006 residents in major Russian cities, supplemented by data from the International Youth Festival.
PhishIntel: Open-Source Python and Go Tool Automates Web Reconnaissance for OSINT Analysts
PhishIntel is a lightweight reconnaissance utility written primarily in Python with performance-critical components in Go. The tool launches via a single command and presents an interactive menu that lets analysts select language and modules. Its core crawler recursively visits a configurable number of pages at a chosen depth and concurrency level, harvesting Russian and US-format phone numbers, emails, cryptocurrency wallets, and physical addresses. Additional modules collect domain metadata including DNS records, RDAP, WHOIS, TLS certificates, redirect chains, and external JavaScript and API endpoints. A limited TCP port scan performed by the Go binary identifies listening services and basic technology fingerprints without replacing full port scanners such as Nmap. Results are exported in both structured JSON and human-readable HTML formats, while an optional .env file allows customization of user agents and proxies.
Smart Engines and Intek Deliver Russian Scanners for AI-Powered Document Verification and Fraud Prevention
Smart Engines has partnered with GK Intek to release a line of Russian software-hardware scanner complexes that combine automatic data entry with multi-spectral authenticity checks for passports and other identity documents. The systems support documents from Russia, CIS countries and 235 jurisdictions worldwide, analyzing optical, ultraviolet and infrared images plus RFID chips to detect forgeries including photo swaps and digital edits. Deployments at Genbank and Severgazbank have reduced manual passport entry to fractions of a second while blocking fraudulent documents in real time without sending images outside the organization. The same technology powers border control modules at Sheremetyevo, Vnukovo and Koltsovo airports inside the Sapsan system, cutting passenger processing time to 40 seconds. Additional use cases include automatic pass issuance at transport hubs, electronic signature issuance by the Federal Tax Service and a pharmacy robot in Minsk that verifies age-restricted sales. A newly patented U.S. method detects hidden digital manipulations by analyzing image provenance rather than content, while another patent identifies optically variable devices to spot copies and replicas.
Why Isolated Bank Scoring Fails to Prevent Counterparty Default: OSINT Case Study of a Scam Deal
A Russian social entrepreneur lost nearly one million rubles after relying on a major bank's automated compliance system that assigned a green reliability score to a fraudulent supplier. The supplier, registered as a transport company, received 500,000 rubles in targeted government grant funds for IT equipment but delivered only 15,000 rubles worth of goods before disappearing. Retrospective analysis using public Russian registries revealed multiple red flags missed by the isolated scoring algorithm, including mismatched OKVED codes, missing financial statements for three years, and clear ownership links to a bankrupt IT firm. The beneficial owner later faced tax blacklist entry and three-year disqualification after the transport entity was liquidated as inactive. The case demonstrates that single-entity automated checks create a false sense of security by ignoring beneficiary networks and operational logic. Manual verification through official sources such as EGRUL, GIR BO, and arbitration court records proved far more effective at exposing the scheme.