In the squo patents allow “prophetic experiments” that scientists believe to be real examples because they are not labeled. This causes misinformation to spread because scientists are not aware they are hypothetical results

In most contexts, making up data is forbidden - considered fraudulent, even immoral. Not so in patents. Patents often contain experimental data, and it is perfectly acceptable for these experiments to be entirely fictional. These so-called “prophetic examples” are not only explicitly permitted by both the Patent Office and federal courts, but are considered equivalent to factual data in patent doctrine. Though prophetic examples are thought to be common, there are no in-depth studies of the practice, nor any explanation for why fictional data are allowed in patents. Here, I provide the first historical, theoretical, and empirical analysis of prophetic examples. I collect and analyze a novel dataset of over 2 million U.S. patents and applications from the biology and chemistry industries. I find that at least 17% of experiments in this population are fictional. Through both empirical and theoretical analyses, I weigh the potential costs and benefits of prophetic examples and find that the costs prevail. Prophetic examples could be beneficial if they help patentees; but I find little evidence that they do so, even in the specific situations in which they should be the most useful. Instead, prophetic examples likely hinder innovation because they prevent others from conducting their own experiments – even after the patent has expired and even if the prophetic example is incorrect. Prophetic examples also hopelessly confuse scientists – a shocking 99% of scientific articles incorrectly cite prophetic examples as if they contained factual information – which means that made-up results from patents contaminate the scientific literature.

To be justifiable, prophetic examples must help patentees, but it is not enough for them to merely help patentees. The underlying assumption in the case for prophetic examples is that they help patentees in a way that is not harmful to the patent system more broadly. This assumption has two main components, each of which is discussed below.

1. Chilling Downstream Research

Prophetic examples might help patentees file earlier, broader patents. However, it is far from clear that these broader, earlier filed patents are actually socially beneficial. Broad, early filed patents are supported by adherents of the “prospect” theory of patent law who argue that such patents allows patentees to coordinate technological development in that area. This prevents wasteful races to invent and reduces transaction costs during downstream development. These are all potential benefits of prophetic examples.

However, some scholars worry that overly broad patents reduce competition and block downstream innovation, and that early-filed patents reflect less developed inventions and therefore lead to patents that are more vague, useless, or, if useful, are never commercialized. The problem with such patents is that they might effectively prevent others from working in the areas surrounding the patent. First, while the patent is in force, others cannot do their own experiments in the area covered by the patent - even if the prophetic examples do not work. Even after the patent has expired or been abandoned, it might still chill research in that area because others cannot get a patent on an invention disclosed in or rendered obvious by a prophetic example. Patents are given only for inventions that are new and nonobvious, therefore material that is disclosed, and everything obvious based on that disclosed material, is no longer patentable. This is true even if the subject of the disclosure was never physically created. We know that companies make strategic disclosures in their patents for the express purpose of preventing competitors from obtaining patents. Prophetic examples may be one form of such disclosure.

For instance, in Ex Parte Botond Banfi, the inventors sought to patent the use of iodide to treat microbial diseases. The PTO rejected the application on the grounds that the invention was not new because it had been disclosed in a prior patent. The prior patent had indeed disclosed use of iodide, but in a prophetic example describing the treatment of asthma (which is not a microbial disease). The prophetic example is:

A 45 year old female with a history of severe asthma with a morning peak flow of less than 3 l/sec is treated with…iodide in an aerosol formulation, 2 mg three times daily continuously. After a week of treatment the peak flow improves to 6 l/sec.

Although the example was both prophetic and did not actually involve a microbial disease, the court reasoned that it inherently disclosed use of iodide to kill microbes. If someone had used the technique, it would have incidentally resulted in the removal of microbes from the throat, even though that was not the main purpose of the treatment.

Though there is no evidence that this example was included for the purpose of defensive disclosure, the example shows how use of a prophetic example can prevent patenting in a wide area around the patent. The prior patentee never tried using iodide to treat asthma (indeed, it is not clear that the technique would work), but the patent effectively prevented others from getting later patents on iodide to treat completely different respiratory illnesses. Innovators are scared away from research in areas near prophetic examples either because they believe that someone has already tried the technique or because they worry that they will not be able to get a patent themselves. Any defense of prophetic examples must balance their benefits to patentees against this potential problem.

2. Inaccurate and Misleading

The second potential cost of prophetic examples lies in their ability to satisfy the enablement and written description doctrines. As part of the patent disclosure, prophetic examples need to both teach other scientists how to make and use the invention and help inventors prove possession of the invention. It is only intellectually coherent to allow prophetic examples to serve these functions if they are actually understood by scientists and if they are accurate predictions.

To illustrate, if prophetic examples are used to teach scientists how to make and use an invention, then they must in fact be able to do so. If prophetic examples describe protocols that are entirely incorrect, then the patent reader cannot rely on them for instruction on how to make the invention. Similarly, if scientists are confused or misled by prophetic examples, then they do not actually teach scientists anything. Further, if prophetic examples are used to prove that the inventor had possession of the invention they must again be accurate predictions. If the inventor’s predictions are incorrect, it is doubtful that the inventor actually had possession of the invention described by those same predictions. In addition, once a patent is granted, prophetic examples are presumed to be accurate. The presumption is only reasonable if prophetic examples are in fact likely to be accurate.

At stake is not only whether patent doctrine is satisfied. The patent’s disclosure is also supposed to promote innovation. The standard explanation is that scientists get new technical information from patents and then use that knowledge to improve the technology or make their own inventive leaps. If the information in patents is confusing to these scientists or is simply inaccurate, then it is much harder for patents to promote innovation through disclosure.