In an era where the lines between reality and simulation are increasingly blurred by generative AI, a new breed of cyber threat has emerged, targeting not just Fortune 500 companies or government infrastructure, but the very fabric of our personal lives. Enter Savi Security, a startup founded by tech industry veterans Patrick and Ryan Coughlin, which aims to arm everyday consumers with the same level of sophisticated protection previously reserved for enterprise-grade defense.
With $7 million in seed funding in their coffers—led by Acrew Capital with support from Magnify Ventures, TTCER, and Resolute Ventures—the brothers are officially launching their iOS and Android application this Tuesday. Their mission: to intercept and neutralize the increasingly convincing AI-generated scams that are currently flooding our text messages, email inboxes, and phone lines.
The Genesis of a Security Startup
The impetus for Savi Security was not born in a sterile boardroom, but from a harrowing, deeply personal incident that struck the Coughlin family roughly two years ago. Patrick Coughlin, who was then serving as the Senior Vice President of security products at Cisco—following the $82 million acquisition of his cloud security startup, TruSTAR, by Splunk—found himself on the other side of a terrifying digital assault.
His mother received a phone call that, by all conventional measures, should have been impossible to dismiss. The caller ID displayed her daughter’s phone number. When she answered, she heard a voice that was unmistakably her daughter’s: “Mom, they’ve got me.” A blood-curdling scream followed, punctuated by the daughter’s voice pleading, “You’ve got to do what they tell you.” A male voice then took control of the call, demanding $1,200 and threatening the daughter’s life in the parking lot of a local Walmart.
The scammer had perfectly spoofed the daughter’s voice, her phone number, and even cited specific, verifiable details about the local retail environment the mother frequented. While the mother eventually kept her composure, contacted her daughter, and discovered the entire event was a synthetic fabrication, the psychological impact was profound.
"What I was thinking, after calming my mom down, is: What has fundamentally changed in the underlying cybercriminal economy?" Patrick recalls. "We are now able to leverage the same kind of sophistication that I had seen pointed at government agencies and Fortune 500 companies, and we’re deploying that sophistication at the consumer."
The Democratization of Deception
The catalyst for this shift is, predictably, the rapid proliferation of Large Language Models (LLMs) and generative AI tools. Before the advent of these technologies, executing a high-fidelity "imposter scam" required significant manual labor: extensive reconnaissance on the target, professional-grade voice synthesis software, and a coordinated team. It was a high-effort, high-risk endeavor that was generally only profitable when targeting high-net-worth individuals or large institutions.
Today, the barriers to entry have been dismantled. A scammer can clone a person’s voice using as little as three seconds of audio, easily harvested from a social media post of a parent narrating a kid’s football game or a public video. The cost to perpetrate these scams has plummeted toward zero, while the potential reach has expanded to the entire population.
"We’re creating fraudsters because we’re bringing down the barrier of deceiving people," Patrick Coughlin notes. "Not only do we have the organized criminals and the syndicates behind this, but everyday people are sort of being tempted into playing fraud."
Supporting Data: The Rising Tide of Imposter Scams
The scale of this crisis is corroborated by federal data. Last month, the Federal Trade Commission (FTC) reported that in 2025 alone, victims of imposter scams reported losses totaling $3.5 billion—a staggering threefold increase compared to 2020.
While historical trends suggested that older Americans were the primary targets for such swindles, the digital landscape is shifting. Research published in 2025 by cybersecurity firm Malwarebytes highlights that Gen Z, despite being "digital natives," is increasingly vulnerable. Gen Z is targeted by text-based scams more frequently than any other demographic, and they report falling victim to these deceptions approximately 25% of the time. This data underscores that the threat is no longer confined to the technologically illiterate; it is a universal problem that requires a technological, rather than merely educational, solution.
Testing the Defense: The "Scam Wise" Proof of Concept
Before going to market, the Coughlin brothers needed to refine their detection models. They launched a free, anonymous web tool called "Scam Wise," which allows users to upload suspicious emails, texts, or photos to be analyzed for fraudulent markers.
The response was overwhelming. In the four months following its launch, Scam Wise processed over 50,000 submissions, with volume growing by approximately 10,000 submissions each week. This provided the brothers with a massive, real-world dataset to train their AI models. The startup currently utilizes Google’s Gemini as its primary engine, but they have engineered their software on an "AI gateway" architecture, allowing them to hot-swap models—such as specialized voice-detection engines—as the situation dictates.
A New Paradigm: Real-Time Intervention
Savi Security’s new mobile app aims to do more than just filter spam; it offers a feature that sets it apart from traditional antivirus software: live-call monitoring.
During a suspected fraudulent phone conversation, a user can opt to "bridge in" the app’s live agent—an AI-driven monitor that listens for behavioral cues and "tells" indicative of a grift. If the AI detects the pattern of a scam, it can alert the user in real-time, effectively serving as a bodyguard for the duration of the call.
This, combined with the app’s ability to screen texts and voicemails, positions Savi as a new generation of cybersecurity. Unlike traditional tools that look for known malicious file signatures or blacklisted domains, Savi is designed to identify the intent and the behavior of the attacker.
The Economic Model: Family-First Protection
In a nod to the fact that cyber-risk is often a family affair, Savi Security has adopted a unique pricing model. For $8 per month (or a discounted $63 per year), a single subscription covers an entire family with no cap on the number of users. The primary account holder can add children, spouses, parents, or elderly relatives who may require additional tech support. This reflects the reality that one vulnerable family member often serves as the "in" for a broader criminal network to compromise an entire household’s assets.
Future Implications: The Arms Race Continues
The launch of Savi Security marks a critical turning point in the consumer cybersecurity industry. For years, the industry has focused on protecting hardware and data storage. Now, the focus must shift to protecting the communication itself.
The implications are clear: as criminals continue to exploit AI to create "human-sounding" and "human-behaving" threats, the only effective defense is to employ a "counter-AI" that operates at the same speed and with the same contextual understanding.
As the Coughlin brothers have observed, we have reached a point where the sophistication of state-level cyber warfare is being repackaged as a service for petty criminals. To combat this, security can no longer be a reactive, static process. It must be dynamic, ambient, and, above all, as intelligent as the threats it seeks to prevent. By bringing high-level security expertise to the smartphone in the average person’s pocket, Savi Security is attempting to tilt the balance of power back in favor of the individual, ensuring that the "blood-curdling scream" on the other end of the line is a sound that fewer families will ever have to fear.
