What I think we actually learn from the case is that the Court wants to do justice and felt that CETUS, being privately funded, and needing this assignment to be upheld was the more deserving because the Stanford PhD would not have been given access to CETUS and thus could not have made his invention had he not agreed to assign inventions to CETUS.

I understand and agree that the suit went up on the Bayh-Dole issue because Stanford thought it had a loser on the assignment issue and so needed to get an automatic assignment to fight the automatic assignment in the CETUS agreement. The problem I have with Stanford is that it should have realized that was also a loser.

The case is of more widespread significance because it necessarily holds that a first to invent inventor has an assignable property right in the invention, as otherwise the Stanford PhD could not have assigned the invention.

If inventors own an assignable property right to their invention upon creation of the invention, then a first to invent inventor owns the invention and a second inventor to invent does not own the invention, or we have two independent owners of the invention. However, we know that the founders meant the first inventor to own the right, and for a second to invent not to own any such right, as they specifically put into the first patent law a right to invalidate any patent on the grounds that the patentee was not the true and first inventor. So, the AIA takes the property of the first to invent and gives it to the first to file (if different), and that may violate the 5th amendments Takings clause (i.e. that no person shall be deprived of property without just compensation). Thus, to be constitutional, the law must be interpreted to read a “just compensation” requirement into the law or be held unconstitutional so the first to invent is not deprived of his property without just compensation. Since the AIA does not provide for such compensation, either it is unconstitutional or the Government must come up with compensation. At least that is the test case to be pursued the first time an important enough invention is taken from a first to invent and given to a later to invent but first to file inventor.

I would caution on reasoning about “employment agreement” . Stanford had a policy that permitted consulting and stated that inventors owned inventions “whenever possible”–even when made with Stanford facilities. Again, noted by the court. Stanford officials encouraged the consulting, and expected to benefit from it, and the employee notified those officials of the deal. All this before the employee invented. There is nothing in the federal funding agreement or Bayh-Dole that requires the employee to assign to Stanford. Thus, there is also nothing in Stanford IP policy at the time –essentially the “employment agreement” for inventions–that required the employee to assign to the university. “Whenever possible” applies and includes the deal the employee entered into with Cetus.

With regard to why the suit in the first place, Stanford relied on a claim that Bayh-Dole entitled them to exclusive title and voided any competing commitments of title. It was only when they lost on appeal that the question of Bayh-Dole became the focus. There, they and a lot of other universities discovered they were very wrong in their representation of Bayh-Dole.

I do not have an explanation for why Stanford could allow consulting, allow the deal they did, and allow the company to rely on that deal, and then decide not to “accept” it.

IF it will work this way could be a godsend. George

A Stanford prior present assignment would not have changed the waiver–just the method of assignment should an invention be within Stanford’s scope of claim.. Stanford have to require a present assignment that operates even if Stanford later waives that requirement and then after a company relies on the waiver to give Stanford access to its core technology, Stanford then could regret the waiver, take assignment, and sue the company for a quarter billion dollars. “We will let you commit to assign so long as we can void the assignment later at our discretion.”

Stanford knew of the assignment to Cetus/Roche. It could not be a qualified later purchaser under 35 USC 261. It had to find a way to void the assignment.
For that it came up with a reading of Bayh-Dole that argued that federal funding vested ownership outright in federal contractors. Roche argued that such a reading of the law, applied after the fact to a private contract, would constitute a taking from Roche without due process or compensation. The Supreme Court rejected the Stanford argument.