Security is considered a right that everyone deserves. More than half of the world's population using smartphones has to be extra cautious about the safety of smartphones as one of the most important tools for smartphone usage. Biometric authentication has been a long time since Apple's iPhone 5S was one of the first devices to implement technology that is now considered to be quite ordinary. However, manufacturers have made many innovations when it is considered that Samsung's Iris browser and Apple's Face ID are role models for today's biometric security.
The smartphone market has never slowed down and innovation continues to come only from unexpected locations. The latest smartphones tend to have a screen sensor that gives them a futuristic look. Fingerprint scanners are still considered the safest way to encrypt information, but may change soon. Research from New York University and the University of Michigan proves it with an AI software capable of bypassing fingerprint security by adapting and improving existing fingerprints until it decodes the code. AI is called oluş DeepMasterPrints “and consists of two operating units. The first unit generates a fingerprint 2D image while the second unit is different from the others, allowing the software to know whether the required print matches the generated print.
On the other hand, most smartphones do not tend to catch all fingerprints; this further increases the chance of bypassing AI. The researchers tested three different levels of security and achieved a successful success rate of 76% from a 1.6% success rate. This means that if 100 people use the safest biometric scanner, the data of at least 12 will be exposed.
When Apple first announced the fingerprint scanner was removed, it wasn't sure if it was a right step to do many things, but they might have a point by looking at the statistics. Apple claimed that it had a more secure encryption level by offering the untouched Apple ID, which would not easily be fooled even on 3D models. Therefore, fingerprint encryption has not been proven to be as secure as people and some time has been spent to find a legal solution. Face recognition AI is still quite new to the market and you may soon know that security manufacturers are living up to the level of presentation.