Savi aims to protect consumers from real-life AI scams such as kidnappers demanding ransom

Brothers Patrick and Ryan Coughlin, who each have impressive careers in the technology industry (Patrick worked in national cyber defense, Splunk and Cisco; Ryan worked in consumer products at Apple and Spotify), have launched a new kind of security startup.

Safe Security It seeks to protect everyday people from a new crop of incredibly compelling AI-generated scams, whether directed via text messages, emails or phone calls.

The company just raised $7 million in seed funding, and will launch its app for iPhone and Android on Tuesday. The round was led by Acrew Capital, with participation from Magnify Ventures, TTCER and Resolute Ventures.

The inspiration for the company came from a terrible accident that happened to the founders’ mother.

About two years ago, Patrick Coughlin’s mother called him, distraught, saying she had just received a phone call from a man telling her he had kidnapped Coughlin’s sister. He was senior vice president of security products at Cisco at the time. (It got there after Splunk bought its cloud security startup TruSTAR for $82 million in May 2021. In 2024, Cisco bought Splunk.)

Coughlin recounted that her cell phone rang with her daughter’s caller ID. During that call, “she thinks she hears my sister’s voice saying, ‘Mom, they’ve got me.'” I heard a terrible scream, and then my sister said: “You have to do what they tell you.” Then a man comes on the phone and says, ‘If you don’t pay us $1,200 now, we’ll kill your daughter in the parking lot of your local Walmart.’

The scammer carefully spoofed Coughlin’s sister’s number and voice and indicated the location of a Walmart store she frequented.

Fortunately, the mother kept her wits about her, called her daughter, and discovered that she was fine. The kidnapping was an AI-generated scam.

Coughlin, like his mother, was shaken.

“What I was thinking about after comforting my mother was: What has fundamentally changed in the basic cybercrime economy that we are now able to leverage the same kind of sophistication that I saw directed at government agencies, and then later at Fortune 500 companies? And now we’re applying that sophistication to the consumer?”

The answer, of course, is cheap and powerful MBAs and other generative AI tools.

Before AI, pursuing such gains for consumers was not financially worth it. It would require in-depth research into the target, technology to fake votes, etc. Such attacks primarily targeted only those with deep pockets, such as corporations or governments, as was the technology used to defend against them.

“There’s something happening now to consumers with AI falling into the hands of cybercriminals,” Coughlin says. The costs of committing such frauds are minimal, and research materials are readily available.

“You can reproduce audio from three seconds of audio, from a publicly available post on social media. So we all have traces of things that are in the ether – like where we talk or narrate, comment on a children’s soccer match being videotaped, put on Facebook.”

Federal Trade Commission He said Last month, people who reported online crimes collectively lost $3.5 billion to phishing scams in 2025, triple the amount in 2020. While the majority of people who report such scams are older Americans, some research suggests that Generation Z is also highly at risk. Research from 2025 by Malwarebytes, a maker of anti-virus and anti-malware tools, reported that Generation Z was more often targeted by text scams than other generations, and It fell for them about 25% of the time.

The Coughlin brothers’ idea was to develop a tool for real-time intervention.

They tested their idea, and the AI ​​fraud detection model they were building, by launching a free website called Scam wise. It is anonymous, no registration required. Simply upload any suspicious texts, images or emails, and Scam Wise will determine if they are likely to be fake.

“We launched this about four months ago. We’ve had 50,000 orders, and it’s now growing every week by about 10,000 orders or more,” Coughlin said.

Scam Wise has proven to be a source of available data to help train Savi’s AI model to detect scams. The startup currently uses mostly Google’s Gemini software, but has built its software on AI Gateway, allowing it to leverage other AI models as needed, such as options for voice detection.

Savi on Tuesday launched a paid product, an iOS and Android app for consumers, that can scan texts, voicemails and incoming calls for scams.

Features like this are available in a lot of different products (like Malwarebytes), but the most impressive feature in Savi is live call monitoring.

During a suspicious phone conversation, the user can choose to add the app’s live agent as a listener. Savi listens to behavioral conversations that can determine if a situation is a scam while the call is in progress.

Savi’s fees are also a bit unusual. It charges $8 per month, discounted to $63 per year, to cover an entire family, and doesn’t set any cap on the number of users. So, one plan can cover a person’s children, spouse, parents, and that uncle who always seems to need tech support. Or anyone else the primary account holder would like to add and provide administrative support to.

AI has changed “how easy it is to be a fraud,” Coughlin said. “We create fraudsters because we remove the barrier to scamming people. So not only do we have organized criminals and syndicates behind it, but ordinary people are tempted to engage in fraud.”

Savi Security’s answer is like a new generation of antivirus-like software: one that uses real-time artificial intelligence just like the bad guys do.