This is false.
Put another way, the very fact that we run such massive tests on them before deploying them is itself evidence that the wrong ones can be very dangerous. (Mostly, as I understand it, "very dangerous" to a small set of people who will have very large reactions, even if most people are fine. The ones that are very dangerous to everyone never gets past the initial phases at all, of course.) We're not doing that because vaccines are all perfectly safe under all conditions, but we just want to hold them back for a year for no reason. We're doing it because we need to be very careful with them.
Though I'd also observe a non-trivial component of that danger is also because we're injecting them. Bypassing the body's co-evolved with the environment's defenses is not to be done lightly. This sort of inhaled vaccine is likely to be much less dangerous. I won't say "safe", but less dangerous; the body already has to deal with essentially arbitrary small amino acid sequences being inhaled all the time. I'd still be a bit worried about sythesizing things that we haven't co-evolved with.
(If you don't know what co-evolution is, I'd suggesting taking a moment to learn about it: https://www.britannica.com/science/coevolution It's a very important part of understanding the disease landscape holistically.)