Intel Hit by a $949 Million US Verdict

A federal jury in Texas handed out a decree for Intel Corp on Tuesday. The verdict required the semiconductor company Intel to pay $948.8 million to VLSI Technology as compensation for the infringed VLSI’s patented computer chips.

During the six-day trial, VLSI claimed that Intel’s Skylake and Cascade Lake microprocessors had violated the patent that covered improvements to data processing.

On the other hand, Intel plans to appeal this verdict, saying that it “strongly disagrees,” further adding that the ruling is “one example of many that shows the U.S. patent system is in urgent need of reform.”

The law firm of VLSI refused to comment on the matter.

In a separate ruling over different patents chip last month, VLSI won around $2.2 billion against Intel, to which the semiconductor company appealed. However, the following month, VLSI lost another patent trial against Intel.

According to a VLSI attorney, Intel’s chips infringement causes “millions and millions of infringements per second.”

Meanwhile, a lawyer from the Californian-based Intel noted that the tech giant developed and innovated their chips independently, adding that their modern microprocessors would not be compatible with the outdated technology of VLSI.

Separately, VLSI has two other pending patent cases against Intel in Delaware and Northern California, with the latter set to begin in 2024.

Intel Releases Real-time Deepfake Detector

On Monday, Intel introduced the world’s first real-time deepfake detector called FakeCatcher, which could detect fake videos with an accuracy rate of 96% and return results within milliseconds.

In recent times, deepfake videos are becoming a threat as they can cause negative consequences and harm, further undermining trust in media.

Senior staff research scientist in Intel, Ilke Demir, noted, “Deepfake videos are everywhere now. You have probably already seen them; videos of celebrities doing or saying things they never actually did”

According to Intel’s post, unlike most available detector that examines raw data to search for signs of inauthenticity, FakeCatcher instead looks for clues by evaluating subtle “blood flow” from all over the face and then translating it into spatiotemporal maps. By combining these with deep learning, it became possible to detect the authenticity of a video in an instant.

With FakeCatcher, Intel believes that users would be able to distinguish what is real or fake, thus restoring the public’s trust in media.