When Tarini Padmanabhuni’s grandfather received a call from a voice that sounded exactly like his brother, he believed the ransom demand was real. The voice was a deepfake, and the family lost money. “What stayed with me wasn’t the money,” Padmanabhuni says, “it was that he had no way of telling.” That incident, about two years ago, led her to found DetectifAI, a San Francisco-based startup building on-device AI voice detection so that phone users can instantly know if they’re hearing a synthetic voice.

The problem is significant and growing. According to the FBI, Americans lost close to $900 million to AI-driven scams last year, up 24% from 2024, with people 60 and older losing twice as much as those aged 50 to 59. Competition exists in deepfake voice detection—companies like Reality Defender, Pindrop, Resemble AI, Microsoft Azure AI Content Safety, and Nuance all offer products. But Padmanabhuni argues that these detection products run in the cloud on remote servers, making it impossible for phone manufacturers to build them directly into devices. That leaves the person being targeted with little defense in the moment.

DetectifAI‘s approach is different. Instead of shrinking large cloud models to fit on a phone, the company designs compact AI models from the start that are small enough to run inside a smartphone’s operating system. The goal is to deliver an instant verdict on whether a voice is AI-generated during calls, in voice messages, and in other audio, with the audio never leaving the device. This privacy-preserving, on-device architecture is the core differentiator.

The business model targets phone manufacturers first. DetectifAI licenses its software development kit (SDK) so that detection ships as a built-in feature of the phone’s operating system. Padmanabhuni draws an analogy to AT&T’s exclusive deal with the original iPhone: the first phone maker to ship DetectifAI gains an edge over competitors, and deepfake detection becomes a standard spec like camera resolution. A secondary revenue stream will come from licensing the technology to businesses and fraud-prevention firms.

The startup already has early revenue. According to Padmanabhuni, DetectifAI handles more than 100,000 calls a month for financial institutions in India. Those calls are placed by AI voice agents that handle debt collections and follow up on loan documents, with deepfake detection and speaker verification on every call. She declined to name customers due to confidentiality agreements.

Padmanabhuni’s background is unconventional. She says she began working in machine learning at age 12 and later studied cyber-physical systems at Manipal Institute of Technology in India, where she became the youngest team lead of what she describes as India’s first driverless racecar division in Formula Student, an international student engineering competition.

Asked about the most rewarding moment so far, she points to a small WhatsApp beta test in which users forward suspicious voice notes and get back an assessment of whether they’re real. One tester, whose own relatives had been scammed, called to say they would pay for it without hesitation. “My grandfather didn’t have that,” she says.

DetectifAI has raised a small seed amount from investors Josh Constine (formerly an editor at TechCrunch) and Manohar Kamath, a principal at consulting services firm KM Growth. The startup is also one of the companies vetted by TechCrunch’s editorial team to compete in the Startup Battlefield competition at TechCrunch Disrupt, October 13–15 in downtown San Francisco.