Mavenir Launches NetAIShield: AI-Native Fraud Protection for Voice, Messaging and Data Operator Services

A multi-layered AI defense platform that correlates threats across every network domain while protecting subscribers in real time.

Mavenir, the software company building cloud-native, AI-by-design mobile networks, announced NetAIShield, its next-generation AI-native fraud protection framework designed to secure telecom networks and subscribers against an increasingly sophisticated fraud landscape.

Modern telecom fraud has evolved beyond isolated attacks. Cybercriminals now orchestrate coordinated campaigns across voice, messaging, data, and OTT channels, impersonating telecom operators, government agencies, and financial institutions while leveraging deepfake audio, synthetic identities, SIM swap, SIM box, and AI-powered social engineering techniques. Traditional fraud solutions, built to detect threats within individual network domains, provide only a fragmented view, allowing increasingly complex attacks to evade detection.

NetAIShield responds to this challenge with a multi-layered AI defense architecture built to correlate network telemetry, subscriber behavior, signaling events, messaging intelligence, and threat indicators into a unified, real-time view of fraud. Designed to work alongside Mavenir’s existing security portfolio or as an overlay for third-party fraud detection solutions, NetAIShield helps enable operators to detect, prioritize, and neutralize sophisticated attacks including the ones utilizing SIM farms and Application Farms with minimal disruption to existing infrastructure.

A key innovation of NetAIShield is AI-Protected Communications, which extends fraud protection beyond the network to individual subscribers. While network-layer security protects the population by identifying and blocking fraudulent traffic before calls are connected, the most damaging scams increasingly occur during live conversations. AI-Protected Communications helps fill this gap with a personal AI assistant that can intelligently screen unknown callers, monitor conversations for indicators of social engineering, deepfake voice impersonation, financial fraud, and other suspicious behavior, and intervene in real time when risks are detected.

Operating transparently on existing devices without requiring new hardware, the capability works on a consent basis and supports regulatory compliance by informing callers that they are interacting with an AI screening assistant. Trusted contacts can bypass screening, while unknown callers are authenticated before reaching the subscriber. For ongoing protection, the AI assistant can remain on the call, providing real-time risk assessment, warning subscribers through whisper warnings, or helping terminate or redirect fraudulent conversations when necessary. The platform continuously learns from emerging attack patterns, enabling protection that evolves as fraud tactics change.

Built on Mavenir’s proven security portfolio, which is already deployed across more than 70 Tier-1 operator networks protecting over one billion subscribers worldwide, NetAIShield represents the next evolution of AI-native telecom security, combining network-wide intelligence with personalized, real-time subscriber protection to defend against the rapidly growing threat of AI-powered fraud. NetAIShield in combination with SpamShield, CallShield, and LinkShield enables mobile operators to create safe mobile experience for safe messaging, safe calling and safe browsing.

Ilia Abramov, VP & General Manager, Security Business Unit, Mavenir: “Fraud has become agentic, multi-modal, and deeply personal. With NetAIShield and its AI Call Screening agent, Mavenir delivers an intelligence layer that sees threat signals and protects subscribers in real time.”