We Are San Francisco

San Francisco Founder Launched AI Voice Detection Tool

The startup developed on-device software to help stop deepfake scams that cost Americans $900 million last year.

Updated on Sept. 28, 2026 in Artificial Intelligence

A close-up view of intricate gold-colored microchip circuitry on a silicon wafer, representing advanced AI security technology.
San Francisco-based startup DetectifAI has launched a new on-device software tool designed to identify and block AI-generated voice scams on smartphone platforms. AI Illustration. Upload story photo >

Live Poll

Do you feel confident that your phone’s current security measures effectively protect you from AI-generated scams?

Tarini Padmanabhuni has launched the San Francisco startup DetectifAI to protect smartphone users from deceptive AI-generated voice scams. The company developed compact models that run directly on smartphone hardware to identify fraud.

Why it matters

The venture aims to address a growing security threat that saw national losses from AI-driven scams surge by 24% between 2024 and 2025. Padmanabhuni founded the firm following a personal experience with her grandfather being targeted by a deepfake voice scheme.

Americans reported $900 million in losses to AI-driven scams in 2025, marking a 24% increase from 2024. DetectifAI currently processes over 100,000 calls per month for financial institutions in India.

The players

Tarini Padmanabhuni

The San Francisco entrepreneur who founded DetectifAI.

DetectifAI

A San Francisco-based startup that builds on-device software to detect AI-generated voices.

Josh Constine

An investor who provided seed funding to the company.

Manohar Kamath

An investor who provided seed funding to the company.

The details

DetectifAI provides a software development kit that phone manufacturers and businesses can integrate into smartphone operating systems. The startup is currently participating in the TechCrunch Disrupt Startup Battlefield competition in San Francisco. Following this appearance, the firm plans to license its technology to additional fraud-prevention entities as a secondary revenue stream.

Timeline

  1. 2024: Base year for tracking AI-driven scam losses.

  2. 2025: Year in which AI-driven scams resulted in $900 million in losses.

  3. October 13-15, 2026: TechCrunch Disrupt Startup Battlefield competition.

Across the Bay

DetectifAI's launch follows the 24% surge in AI-driven scam losses recorded between 2024 and 2025. The company's focus on on-device software represents a shift in security strategies compared to the traditional, cloud-reliant verification models used by financial institutions.

While the technology is currently being used by financial institutions, future integration into smartphone operating systems could eventually provide residents with local deepfake protection. Keep an eye on local tech updates to see when these security features become available on your device.

The takeaway

As AI scams continue to rise, consumers should remain skeptical of unexpected calls even if they recognize a voice. Follow the results of the TechCrunch Disrupt Startup Battlefield this October to track the progress of new detection technologies.

Further reading

For more developments on local AI research and security, visit the Artificial Intelligence section.

Live Poll

Do you feel confident that your phone’s current security measures effectively protect you from AI-generated scams?