This is not the argument. I don't want to be the "strongest empire ever!!1!" and bitching on the forums because of that. I sincerely believe it's for the better, both from a RP and a gameplay perspective, or I wouldn't have posted this. You have every right to disagree and I will always welcome discussion, but please don't minimize my points mocking them.
My apologies, I should've chosen my words more carefully. My point here, however, remains this: Ascension Perks are a part of the game as it stands right now. The Ascension Paths are an amalgamation of "culture" and "technology", changing how the game works, while the other perks largely offer percentage bonuses to already existing game components. So far so good. As far as I can tell, the game-changing Ascension Perks are the ones most-coveted by people for RP reasons, and the more powerful ones among them also for gameplay reasons. These are: Ecumenopolis (Arcology Project), Ascension Paths (Machine Ascendance, Psionic Ascendance, Biological Ascendance), Megastructures (Voidborne, Master Builders, Galactic Wonders), Machine/Hive Worlds, Colossus Project (mostly for the total war CB), and, to a lesser degree, Enigmatic Engineering and Nihilistic Acquisition. All the other perks fall into the "filler" category right now (with the possible exception of Gaia terraforming, don't know, hard to tell in practice when it's a little broken - and not very impactful at first glance).
To outline my personal bias:
For me personally, I tend to play tall empires (mostly because I don't have time/patience to manage 30+ planets all growing at the same time), and thus end up taking most of the abovementioned. A usual list of Ascension Perks for me is Voidborne, Gene Ascension #1, Arcology Project, Master Builders, Galactic Wonders,Gene Ascension #2, Tech Ascension (filler), Colossus Project/second filler (Mastery of Nature is a common one). With this build I can quite comfortably have a first full megastructure up by 2350 (sooner if I don't have to deal with early aggressive neighbors, I've had them up as early as 2330), ready and supporting my economy to deal with the crisis, and a Ringworld in progress to deal with the looming population problem. At the same time, my tech isn't quite up to repeatables in Engineering at this point. You're proposing to absolutely kill this possibility, neutering 2 (or 3, depending on how you count) of the 6 AP picks, since lategame tech would enable megastructures becoming available no sooner than 2350 (usually), and thus coming fully online no sooner than 2400, far too late for the resource ones to have an impact on being able to deal with the endgame crisis and/or AFEs.
With this in consideration, and the reasoning you give for your proposal taken at face value, the logical conclusion is that you are calling into question the very essence of what (the interesting) Ascension Perks are at the moment. Keep in mind that unless further restrictions are in play, as the game currently stands,
all empires will research
all technologies by the lategame. By making megastructures available via tech, you seek to make them more common rather than something that sets an empire apart from others.
As far as RP goes: it took the Fallen Empires presumably a long time to reach the stage of technological development they're at - and yet, not all of them have megastructures. It stands to reason, then, that as far as current Stellaris lore goes, not all empires obtain the knowledge necessary to build megastructures. Is this logically sound, from an out-of-game science-fiction perspective? Perhaps not (I would be among the first to cry out "why don't they all have multiple Dyson Spheres?!?"). But it is the current ingame lore situation. So as you see, the RP argument ... has some potential issues.
As for the gameplay - it is a question of what megastructures are supposed to represent. The pinnacle of interstellar engineering, and thus on par with the ascension paths? Or the logical conclusion to any technological development? Or, perhaps, "wonder" type buildings? Currently, they are a mix of the first and last options - this wasn't always the case, initially you could build a potentially unlimited amount of any megastructure, but the omnipresence of the Dyson Spheres made for poor gameplay, and thus we are where we are today. If you make the megastructures into the conclusion of certain technological developments, what's to stop a player from pursuing a path that gets them Dyson Spheres and Research Nexuses, then making 1-system vassals that are now also capable of building such; then feeding said vassals with the resources to do it, and re-integrating them again ... for virtually unlimited Dyson Spheres and Research Nexuses?