That's a valid concern. We need to look at any and every possible "loophole" (aka WDTTDT!) and assume they absolutely will do that. If they claim they won't, then it should be extremely easy to tack on iron clad language (with teeth) preventing it with significant penalties if they try. I mean if they'd never do something, they can't possibly oppose protections from doing it.

The other side of this is that they know full well what they intend to do, now and in the medium term. Just because our side may sign an NDA showing some plans or data about something doesn't mean they were told everything.

We can bet that if they can outmaneuver us on scope, they will do so eventually. Sort of a corporate version of the "Cantillion Effect" that we may not see until after we've thrown them into the briar patch.