Brothers Patrick and Ryan Coughlin, each with impressive careers in the technology industry (Patrick worked for National Cyber Defense, Splunk and Cisco, and Ryan worked on consumer products at Apple and Spotify), have launched a new kind of security startup.
Savi Security aims to protect the public from new and incredibly convincing AI-generated scams, whether routed via text, email, or phone call.
The company just raised $7 million in seed funding and plans to release an app for iPhone and Android on Tuesday. The round was led by Acreal Capital with participation from Magnify Ventures, TTCER, and Resolute Ventures.
The company's inspiration came from a horrific incident involving the founder's mother.
About two years ago, Patrick Coughlin's distraught mother called him and told him she had received a call from a man who had kidnapped Coughlin's sister. He was Cisco's senior vice president of security products at the time. (He got there after Splunk acquired his cloud security startup TruSTAR for a reported $82 million in May 2021. Cisco acquired Splunk in 2024.)
Coughlin said her cell phone rang with her daughter's caller ID. During that call, “she thought she heard her sister say, 'Mom, we got arrested.'” A blood-curdling scream was heard, and her sister said: “You have to do what they say.'' Then a man came on the phone and said, “If you don't pay me $1,200 right now, I'm going to kill your daughter in the parking lot of the local Walmart.''''
The scammer used the exact phone number and voice of Coughlin's sister and mentioned the Walmart location she frequents.
Fortunately, the mother was quick-witted and called her daughter to make sure she was okay. This kidnapping was a fraud created by AI.
Coughlin was just as upset as her mother.
“After I calmed my mother down, I was thinking, what has fundamentally changed in the underlying cybercrime economy? That we now have access to the same kind of advanced technology that has been directed at government agencies and then Fortune 500 companies, and now we're deploying that advanced technology to consumers?”
The answer, of course, is cheap and powerful LLM and other generative AI tools.
Before the advent of AI, pursuing such detriment to consumers was not economically worthwhile. This requires careful research on the target and technology to disguise voices. These attacks primarily targeted only the wealthy, such as corporations and governments, as did the technology used to protect them.
“Something is happening to consumers right now because AI is in the hands of cybercriminals,” Coughlin said. The cost of running such a scam has become negligible, and research materials are now easily available.
“You can replicate three seconds of audio from a public social media post. So we all have traces of things that exist in this world, whether it's talking, narrating, videotaping and commenting on your child's football game and posting it on Facebook.”
The FTC announced last month that people who reported online crimes lost a total of $3.5 billion to identity fraud in 2025, triple the amount in 2020. Although the majority of people reporting these scams are older Americans, some research suggests that Gen Z is just as vulnerable. A 2025 study by Malwarebytes, a maker of antivirus and anti-malware tools, reported that Gen Z was targeted more often by text scams than other generations, about 25% of the time.
The Coughlin brothers' idea was to develop a real-time intervention tool.
They launched a free website called Scam Wise to test their idea and the AI fraud detection model they were building. It's anonymous so no registration is required. Simply upload a suspicious text, photo, or email and Scam Wise will determine if it's likely to be fake.
“We started this about four months ago. We've had 50,000 posts so far and are now growing by about 10,000+ posts every week,” Coughlin said.
Scam Wise has proven to be a useful source of real data for training Savi's fraud detection AI models. The startup currently primarily uses Google's Gemini, but because it's building its software on top of AI Gateway, it can leverage other AI models as needed, including options specific to voice detection.
On Tuesday, Savi released a paid product and consumer-facing iOS and Android apps that can screen out incoming texts, voicemails, and fraudulent calls.
Such features are available in a variety of products (such as Malwarebytes), but Savi's most impressive feature is its live call monitoring.
During a suspicious phone conversation, users can choose to add the app's live agent as a listener. During a call, Savi listens for action instructions that can identify whether a situation is discordant.
Savi's prices are also a bit special. Cover your entire family for $8 a month, discounted to $63 a year, with unlimited users. That means you can use one plan to cover your kids, your spouse, your parents, and that uncle who always seems to need tech support. or anyone else the primary account holder would like to add to provide administrative support.
Coughlin said AI has changed the terms of “how easy it is to be a fraudster.” “We're creating fraudsters because we're removing barriers to deceiving people. So it's not just organized criminals and syndicates that are behind this, but everyday people who are kind of seduced into deception.”
Savi Security's answer is a new generation of antivirus-like software that uses AI in real-time just like the bad guys.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.