Keeper Security adds local AI-powered phishing protection to its Browser Extension to block credential theft
Keeper Security has launched an opt-in Phishing Protection feature in the Keeper Browser Extension (version 18.2.0). It uses a local AI engine to detect look-alike domains, spoofed websites and brand impersonation before users enter credentials. The vendor cites the 2026 Verizon Data Breach Investigations Report, which found credential abuse in nearly 40% of breaches, and says attackers increasingly use generative AI to build convincing fake login portals. This is product news with no direct threat or vulnerability.
Detalhes técnicos
- The machine-learning engine runs entirely on the user's device and is built on Keeper's zero-knowledge architecture, so website data is not sent to external servers.
- It looks for typosquatting, suspicious IP addresses, unusual URL structures and unauthorised sites requesting passwords.
- When it flags a page, the extension shows a warning listing the indicators detected. Users can dismiss the warning, disable Keeper on that site, manage blocked sites and restore autofill.
Remediação
- The feature is disabled by default. Enterprise administrators can enable and manage it centrally in the Keeper Admin Console, and individual users can configure it in the extension.
- It is meant as an additional layer against phishing that evades email security gateways and web filtering.
Linha do tempo
-
Keeper Security adds local AI-powered phishing protection to its Browser Extension to block credential theft
ler a análise desta fonte ocultar
Keeper Security has launched an opt-in Phishing Protection feature in the Keeper Browser Extension (version 18.2.0). It uses a local AI engine to detect look-alike domains, spoofed websites and brand impersonation before users enter credentials. The vendor cites the 2026 Verizon Data Breach Investigations Report, which found credential abuse in nearly 40% of breaches, and says attackers increasingly use generative AI to build convincing fake login portals. This is product news with no direct threat or vulnerability.
Technical Details
- The machine-learning engine runs entirely on the user's device and is built on Keeper's zero-knowledge architecture, so website data is not sent to external servers.
- It looks for typosquatting, suspicious IP addresses, unusual URL structures and unauthorised sites requesting passwords.
- When it flags a page, the extension shows a warning listing the indicators detected. Users can dismiss the warning, disable Keeper on that site, manage blocked sites and restore autofill.
Remediation
- The feature is disabled by default. Enterprise administrators can enable and manage it centrally in the Keeper Admin Console, and individual users can configure it in the extension.
- It is meant as an additional layer against phishing that evades email security gateways and web filtering.