Update

A new Loop-Hole to invade AI

With the advancement of technology, hackers became more jeopardizing. To deal with them, we have to be one step ahead of them, i.e; to learn new methods of how to attack could be done and further to deal with it. Recently, researchers found a new inimical attack that can even fool AI technologies. It is named as “OPtical ADversarial attack (OPAD), requires just three basic things to perform the drill- a camera, a low-cost projector, and a computer.

About the attack:

OPAD is a cheap and cost-efficient method, which could be used to calculate patterns and further modify the appearance of 3-D objects. For trials, the researchers utilized seen objects by AI which exist. The attack was performed by projecting few calculated patterns over the image. OPAD is capable of targeting 3-D objects in a single shot and can even launch untargeted, targeted, black-box, and white-box attacks as well. The consensual loss caused by OPAD is visible to the users.

More in Depth:

One of the major merits of OPAD is, no physical access to the object is required. It can transform any digital data into real 3-D objects. Although it is limited only to the surface material, can be used to fool self-driving cars, fooling AI-based security cameras, hence tampering with the real data.

Final thought:

AI companies should be aware and stay alert as these potential securities are always there in the outside world seeking a slight chance to slip into your system.

Riya

Riya is a technology enthusiast and an avid researcher. She writes about consumer tech, hacking, and technology consumer issues at TheDigitalHacker.
Back to top button
Close
Close