Two years ago, a local Waco, Texas judge ruled that Intel owed patent owner VLSI Technology $2.18 billion in patent-infringement awards. The VLSI patents cover two technologies: frequency management developed by SigmaTel, which accounted for $1.5 billion of the verdict, and a memory voltage reduction technique from Freescale, which accounted for $675 million. Intel unsuccessfully attempted to overturn the verdict in 2021, and earlier this year, a U.S. patent office tribunal invalidated the two patents in a separate proceeding that could further erode the value of the jury’s findings.
On Monday, a U.S. appeals court threw out the award, overturning one of the largest verdicts in the history of U.S. patent law and slamming Intel for engaging in “patent trolling.” The decision has broad implications for chipmakers and other companies involved in patent disputes. The decision may prompt them to reassess how they approach damages calculations in future cases.
The court’s ruling, which sends the case back to a district judge for a new trial to determine how much Intel owes for infringing the second VLSI patent, primarily concerns VLSI’s damages analysis. Specifically, the court ruled that VLSI’s expert should have revised how much value Intel would lose by using an incorrect model for how much to estimate damages.
During oral arguments, U.S. Circuit Judge Timothy Dyk expressed skepticism about the damages model used by VLSI’s expert. He said that “it seems to me pretty clear” that the expert’s damages model relied on non-infringing products and features of Intel’s devices to make his calculations and that he may have committed a material error. Judge Richard Taranto agreed with Dyk, saying it “seems a safe bet” that the expert’s model was wrong.
The decision also criticized the expert’s reliance on documents that included information that was private and his failure to include evidence that Intel had used its patented technology in prior products. Intel argued that VLSI’s failure to produce more detailed damages estimates made it impossible for the jury to evaluate the accuracy of his conclusions.
The court also ruled that the jury was incorrect to find that Intel’s conduct was willful or reckless, a critical factor in awarding enhanced damages. The company argues that its evidence does not support a finding of willful infringement and that its motions for judgment as a matter of law on these issues should be granted. The case is VLSI Technology v. Intel Corp et al., 2024-2029, in the U.S. Court of Appeals for the Federal Circuit. The court’s full opinion will be published later this week. The court does not usually publish summary and non-precedential appeals decisions. However, the court cannot publish any critical or significant opinions. The court has 13 panels based in various cities across the country. The court typically hears arguments on patent-related cases in a city outside of Washington, D.C.
