A Natural 1 does not necessarily conjure these things into being as you said. That is the mischaracterization of the Cypher System that I find problematic, and I will get to that more below.
Regardless of how you want to characterize it, the truth is that these complications - however plausible - are introduced into the fiction at that moment that action is attempted. If these complications don't associate with the actor attempting the action or perhaps with the target of the action making a countermove, then they are indeed conjured out of thin air. They are plausible within the narrative, and obviously agreeing to play in that style is agreeing to allow the fiction to be morphic and modified on a whim.
So yes they come from the realm of narrative possibility, but that is also out of thin air and indeed it is the missed bow shot that conjures them into being.
So you do not understand what I'm arguing or where I have a problem. The problem I have is that Monte claims that it is better that usually the complication introduced by GM Intrusion be one which is not associated with the actor, or necessarily even with the target, but rather be like random coincidences invoked by some diabolic spirit that hates the number '1' or what you call "a sort of organized chaos effect". But I don't agree that that is necessarily or even often a better approach, I think it one fraught with problems regardless of the system you are playing, and that the problem Monte is trying to solve here is not serious enough to warrant repeatedly using disassociated mechanics or repeatedly improvising antagonistic elements of the fiction.
Now, Monte has quite correctly given the player Narrative currency that he can spend to mitigate against the GM's call, but as I understand it, the Cypher system doesn't implement a separate narrative currency pool but instead the player has to spend XP to counter the DM's call. And if the GM can call on roughly 1 in 20 rolls, then the intrusion should be relatively minor and inoffensive IMO. So yes, it probably shouldn't belittle the PC, but it should probably not mostly be zany introductions to the fiction either.
One way to handle this functionally in combat would be to have the monster make a move, preferably one that the monster could do. So you might attack the orc, roll a 1 and the DM narrates this as the orc warrior catches the blow on your blade, locks swords with you, with its fang mere inches from your face, and with a roar the monster throws you onto your back! Oh noes!! Things of that nature are good combat calls by a GM. Or you could have the monster use the established terrain of the fiction in some manner, bulrushes you and knocks you to the edge of a cliff, takes advantage of your poorly timed swing to pull a brazier of coals down on top of you and the straw on the floor of the room catches fire, knocks your sword spinning out of your hand, etc. The GM should be creative with what the fiction has already established - that the orc is a strong and frenzied attacker, those things already present in the environment, and yes that the player - as the actor in the proposition - sometimes makes a big mistake is momentarily outmatched.
There are lots of ways to handle this well, but the Cypher system doesn't really have the narrative currency you need to build fiction together in a functional way and Monte's games don't normally play well as strictly Nar games because he has such strong Gamist and Simulationist sensibilities. As someone else I heard comment said, "Monte makes games if you want to heroicly knock down the bad guys." I don't think that he's getting peanut butter into his chocolate here, or even that he's dipping his nori in chocolate because fad. I think he's using the wrong technique for the wrong reason.
But it seems as if you are fighting for a cause that does not affect you since you are playing in other game systems with other play styles anyway. So what does an online debate regarding Monte Cook's article on fumbles and the GM Intrusion achieve for you? Why are you worried about what effects these new mechanics would produce in a game system you don't even play or imply that you would enjoy regardless of this mechanic?
Because these are general technique that aren't tied to any one system and there is this fad going on where people think that Nar techniques are just better regardless of how or when or why you use them. And I don't think that's a very thoughtful approach here. Besides which, I enjoy thinking about GMing techniques because regardless of what system I'm playing now, I'm GMing all the time. I've been GMing for more than 30 years and I'm always learning about new approaches and ways of looking at things. Right now I'm 6 years in to 3.X D&D campaign that probably has 2-4 years left in it to complete it's major story arc, but even in D&D 3e you can make calls and improvise new fiction. It's just not a technique you want to do very often because it the system doesn't give the players tools to respond to that. If you find yourself in 3e doing a lot of that, you are probably doing it wrong.
And in my opinion Cypher doesn't support introducing disassociated improvised fiction through the GM Intrusion mechanic very well. XP is too high of a cost to counter it with; it's not fun for the player to pay that cost, which should be fun to 'bid' in a well designed Nar mechanic. Cypher has more than one creative agenda going - it's got a wonderful exploration of setting theme going for example - so don't expect it to play pure Nar.
It's not necessarily a fumble. It's not necessarily a comedy of errors for the player, as Natural 1s sometimes become.
Comedy of errors is being underrated here. Heroes can bumble and fumble and still be big damn heroes (to quote Firefly) and the players can still have fun. In fact, I've had players that signal that they very much enjoy that sort of thing. And if I had a player that signaled strongly that they didn't, I'd try to meet him halfway. And yes, there are ways to handle GM intrusion that aren't comedy of errors or even strongly color of PC incompetence (though there is IMO _always_ going to be some just because the players can see the metamechanic in action), but in general when roll is failed the results should always be strongly and clearly related to the act and the actor. Otherwise you are running into a problem that the players aren't going to feel that they can control the game setting through their actions (and they'll be right) and the universe (and the GM) is perverse and illogical and out to get them. And it's particularly bad to have this happening on a 'roll of a 1' because they know this is happening, and they don't have sufficient tools for saying 'No'.
It's not necessarily a bumbling goof. It's really more an opportunity that invites the GM-as-narrator to provide unforeseen narrative twists. And these are moments that we see in movies, shows, and books. The player can do things well, but sometimes the unexpected happens that lies outside of their competence or foresight.
Sure, but the bumbling goofs also happens in movies, shows, and books and we don't necessarily like the heroes the less for them. And things that are unexpected and lie outside of any possible foresight will happen to PC's anyway, regardless of whether we have GM Intrusion as a mechanic or not, just because the player's themselves aren't omniscient (at least, not in this sort of game, and if you want to talk about a game I wouldn't actually like to play, a game with all the players omniscient would be an example).
You may see this as splitting hairs or a case of semantics. That's fine. But semantics do impact the game, as we have discussed much earlier in this thread, and which was something that you seemed to tacitly agree with.
I am the world's biggest semantic hair splitter. You don't need to convince me semantics are important. I just ask you to try to understand what I'm actually saying and why, which is sometimes difficult I grant you, because often what I'm saying is finely split and often as not I'm arguing because I haven't hit on exactly how to say it well.