Hwang Woo-suk was a Korean scientist who claimed to have developed a breakthrough cloning technique in 2004. But he was fired in 2006 after he admitted his results had been falsified.

But that didn't stop the United States Patent and Trademark Office from issuing him a patent based on the fake research. The New York Times was first to report on Patent 8,647,872, which was awarded on Feb. 11.

This would already be awkward enough for the patent office, but the story actually gets worse. According to Matt Levy, an attorney at the Computer and Communications Industry Association, the patent office initially rejected Hwang's application due to his misconduct. Levy describes what happened next:

But just last year, Hwang submitted an affidavit claiming that he had actually done the things the examiner cited, and giving his expert opinion that they were within the skill of an ordinary person in the art.
One might expect the examiner to respond with peals of laughter followed by throwing the affidavit in the trash. Nope.
The examiner accepted the affidavit and issued the patent.

So is this the patent office's fault? Surprisingly, many legal experts say no. Dan Burk, a legal scholar at the University of California at Irvine, describes it as "not terribly surprising" that an application based on fraudulent research could be approved by the patent office. "The Patent office is not the Food and Drug Administration," Burk argues. "They're not in the business of guaranteeing the safety or efficacy of things that are submitted to them."

James Grimmelmann, a law professor at the University of Maryland, agrees. "The general presumptions run in favor of the inventor," he says. "The initial attitude is one of trust rather than distrust."

But that creates a serious problem, Grimmelmann says: Once a patent has been approved by the patent office, the courts accord it a "presumption of validity." If Hwang were to sue someone for infringing his new patent, the courts would assume the patents were valid unless the defendant could prove otherwise.

That might not be difficult to prove in this case, given the wealth of evidence of improper conduct by the ostensible inventor. But, Grimmelmann wonders, "why should it be the defendant's job to show that the person who lied actually committed fraud?"

And even when the case against a patent is obvious, getting the patent thrown out can be expensive. "The defenses you would raise to validity are costly," Grimmelmann says. "If you're going to argue inequitable conduct, you have to show intent to deceive the patent office."

That's not something you can prove just using the materials in the original patent application. It requires gathering additional evidence about the applicant's actions. Doing that will require "lawyers litigating over privilege and relevance claims" — and charging hundreds of dollars per hour along the way.