For an invention to be made, it has to be recognized as inventive by the inventor. See MPEP 2138. “There must be a contemporaneous recognition and appreciation of the invention for there to be conception.” For university investigators, this means that until you recognize what you have done as inventive, it isn’t. Without that, no conception, and therefore no invention is made.
University IP policies now typically require disclosure of inventions. Some nuttily go so far as to require disclosure of inventions “whether or not patentable.” What’s a non-patentable invention? It can be most anything–whatever someone wants to call “inventive”–even if not new, or useful, or non-obvious. But if an “invention” is not patentable, then what’s the ownership theory? Certainly not patent law. Copyright? Then it’s not a non-patentable invention, it’s a work of authorship. Different policy, different issues. So it’s a total crock to have the wording “whether or not patentable” in a university IP policy. It’s trade secret talk, review of all publications talk. Not appropriate for a university.
There are some organizations that want to review all proposed publications for patentable subject matter. They want to create the conditions that the authors of the paper will recognize and appreciate their work as inventive, in case they hadn’t already. They want to induce an invention.
A good reason why inventors need to recognize an invention is that inventors also matter for establishing what is ordinary skill in the art. This is especially true in university research settings. It’s up to investigators to establish the threshold for this. I argue that the quality of research originated patents depends on it. If inventions are induced, in a sense, what is happening is that the quality of disclosures is moving down the ladder. That runs against a selective organization looking for being known for working with high quality stuff.
Note, this is different from recognizing that something is inventive, and having a discussion about whether it is in anyone’s interest that a patent application is filed. And that is of course distinct from a discussion of what to do with a patent once it issues.
Bayh-Dole does not require inventors to report half inventions or stuff other people might think is an invention. The responsibility is with the investigators to decide if something is an invention, or should be published without patent hubbub. The university’s involvement begins when the investigators report they have an invention.
Notice that no matter what paperwork an investigator signs, whether a promise to assign or a present assignment of future inventions, there’s nothing to assign if the investigator doesn’t recognize and appreciate work as inventive.
This comes up in university investigators consulting with industry. They ask, at what point do I have to disclose to my university that I’ve invented in my consulting? Clearly, when a patent application is filed, that constitutes conception and constructive reduction to practice. At that point, one is on record believing there’s an invention. So that would be a good point to disclose if required to do so by policy.