User behaviour patterns can provide the best key for securing devices like smartphones, but not if our data is still collected on centralized servers.
Your smartphone senses your every move. Today’s smartphones contain accelerometers, gyroscopes, GPS, touchscreen, cameras and other powerful sensors that allow them to map your movements in detail. And as the Internet of Things adds more smart devices, wearables and connected cars to your life, the question becomes – how do we keep users secure?
Machine learning and AI unlock new ways to secure users. The powerful learning capacities of algorithms can be trained on the users of smartphones, other IoT connected devices, or connected cars, to recognize the legitimate user via their behavior patterns. But the privacy of this approach is compromised if the data is stored on centralized servers.
Zighra’s newly patented advances in continuous behavioral biometric authentication could provide the basis for a new paradigm in device security and user identity. The core concept is simple. Observe the behaviour of a user over time, use the observations to determine if that user is attempting access, and complete the whole operation on the device in a decentralized solution.
A machine that watches your every move.
AI and machine learning are applied in Zighra’s technology to build a highly accurate picture of a device user’s unconscious behaviour. The high powered technology in today’s smartphones can constantly monitor user behaviour, then compare it to stored patterns for the device owner.
To avoid the problem of “false positive” security alerts, Zighra monitors a plurality of observable behaviours. A single mismatch between stored and observed patterns may not trigger device security. But a string of mismatches, possibly indicating unauthorised access to the device, can trigger lockdown or other security responses.
Significantly, Zighra’s technology delivers its AI driven solution entirely on-device. This has been the central approach since the company’s founding in 2009, and for a simple reason – privacy. Behavioral biometric authentication offers real security advantages, but they can be easily undermined if that data is still centrally collected and stored.
AI at the edge and on-device.
Zighra’s approach to behavioral biometrics, with its decentralized, on-device architecture, improves upon many of the privacy issues inherent in solutions that rely on centralized servers.
- Zighra’s technology answers the call for a “passwordless future” from the likes of the FIDO Alliance and the World Economic Forum. Passwords are vulnerable to a host of security hacks, including attacks on centralized servers. Continuous behavioral intelligence along with biometric authentication removes the need for passwords.
- The increasing power of smart devices allows AI to be implemented at the edge tier, reducing the communication load with platform and enterprise tiers. Zighra’s behavioral authentication implements AI at the edge for a true on-device experience.
- As sensor based devices become more and more prevalent in our physical lives the expectation of users are moving towards continuous and frictionless authentication. Behavioral biometric authentication meets these user expectations, but to do so while maintaining user privacy is essential for true democratization of this technology.
With Zighra’s technology, identity is entirely determined on the device. No data leaves the device, protecting user privacy, and helping organizations comply with regulations such as the European Union’s General Data Protection Regulation (GDPR) and the California Consumer Privacy Act.