Savi raises $7M seed to combat realistic AI-driven consumer scams
Savi Security has closed a $7 million seed round to launch an AI-powered defense app. The startup, founded by brothers Patrick and Ryan Coughlin, aims to intercept real-time scams by leveraging data from their earlier testing platform, Scamwise.
Savi Security, a startup founded by brothers Patrick and Ryan Coughlin, has raised $7 million in seed funding to launch a mobile application designed to protect consumers from AI-generated scams. The round was led by Acrew Capital, with participation from Magnify Ventures, TTCER, and Resolute Ventures. The company’s app, which launched on iOS and Android this Tuesday, aims to screen texts, voicemails, and incoming calls for fraudulent activity. The founders, who bring experience from roles at Cisco, Splunk, Apple, and Spotify, developed the technology to address a growing consumer threat landscape where generative AI has significantly lowered the cost and technical barrier for cybercriminals to execute sophisticated impersonation attacks.
The impetus for the company stems from a personal incident involving the founders' mother, who was targeted by a convincing AI-generated kidnapping scam that spoofed her daughter’s voice and phone number. Patrick Coughlin, formerly the senior vice president of security products at Cisco, noted that the underlying cybercriminal economy has shifted. Previously, high-sophistication attacks were largely reserved for government or enterprise targets due to the high costs of research and execution. With the proliferation of cheap, powerful large language models, these tactics are now being deployed against individual consumers at scale, making formerly expensive grifts financially viable for attackers.
To train its detection models, the startup utilized data from a free, anonymous web tool called Scamwise, which the founders launched about four months ago. The tool allowed users to upload suspicious texts, emails, or photos for analysis. According to Coughlin, the platform gathered 100,000 submissions by this Tuesday, with volume growing by approximately 10,000 per week. Savi currently utilizes Google’s Gemini as its primary model but has engineered its software on an AI gateway, allowing the system to integrate other models, such as those specialized in voice detection, as required for specific security tasks.
Savi’s commercial strategy focuses on a family-oriented subscription model, charging $8 per month or $63 per year for unlimited users. This approach is intended to allow primary account holders to extend protection to children, spouses, and parents. While similar protective features exist in other security products, Savi differentiates its offering through live-call monitoring. Users can opt to add an AI agent to an active phone conversation, where the software listens for behavioral indicators of fraud in real time. This feature is designed to intervene during the critical moments of a scam attempt.
The market for such tools is expanding alongside the rise in reported fraud. Data from the FTC cited in the report indicates that consumer losses to imposter scams reached $3.5 billion in 2025, a threefold increase from 2020. While older demographics have historically been the primary targets, research from 2025 by Malwarebytes suggests that Gen Z is also highly susceptible, falling for text-based scams approximately 25% of the time. As Savi enters the market, its ability to effectively scale its real-time detection capabilities will be a primary indicator of its potential to capture share in the consumer cybersecurity sector.