155A. The End of Alignment
What exactly IS alignment? This concept is as old as history, but is one of the most mysterious and secretive concepts known to humankind.
Just 11 weeks ago I wrote a very detailed paper on AI Alignment. There I made the argument that all of the concern over whether “AI is aligned or not” might be irrelevant:
The reason I am writing on this subject today is because if there is one subject that screams “The Matrix” louder than any other, it is this concept of “Alignment”. This word is getting a lot of play these days because of AI, but this concept is a very old concept that shapes our world and maintains power. I want to really show what this looks like in human space, because what I’m going to write about in my next paper is going to blow your mind so hard that some preparation would really be helpful.
In AI Alignment I made the case that an aligned AI was just as dangerous as a misaligned AI. Our media is currently programming us to see “Aligned” AI as good and “Misaligned” AI as bad. But at its core, this programming is there to teach you that:
Alignment is good for you.
Let me give a real life example of how this works in the real world. In my new intro article I explained how I got here, and how I was identified as a trusted expert despite not having a PhD. I want to examine that academic scenario through the lens of Alignment so you can understand that your alignment is tested and rewarded/punished regularly, maybe even daily. Your position in life is only marginally due to merit. Alignment is everything. Unfortunately for many, you can be born misaligned.
Now I’m not stupid. When I applied for the PhD program at USC I already knew that this was not a test of merit, it was a test of alignment. Here are the rules they want you to believe determine the success of your academic applications:
GPA/Coursework
Prior Works/Contributions
Academic Referrals
I’d played this game before so I knew this wasn’t really how they evaluated candidates. This wasn’t about merit, it was about alignment at this level of power. And all universities aren’t the same. UC Riverside might have as good or better an academic reputation as USC, but you don’t go to USC for an education. You go to USC to get access to power, which is also why “ivy league” schools are special. They open doors.
So my strategy was to 10X all three of the above official categories. I’m told long ago there was also a “test of character” but we got rid of that a long time ago. We don’t interview or even meet candidates anymore. And I did exactly that, I produced the highest possible GPA, the best possible written work (Sustainable Virtual Economies and Business Models), and I had the top two professors supporting my application as wingmen, both there at USC.
I either wanted to force them to accept me, or force them to admit the criteria were illusory. Because as someone that was born misaligned, I knew the criteria likely also included or were completely replaced by:
Your caste
Your religion
Your ethnicity
Your political orientation
I knew all this back when I was 14 and helped Bob Dornan push the B-2 Bomber through Congress.
So you can see I checked all of the official boxes, and I was top among all candidates in those 3 categories. Normally I wouldn’t have access to the GPA scores of all applicants, but the ladies in the counselling office were sweet on me and offered that information up to me. I wasn’t taking any chances. The problem was I didn’t check any of the real boxes: the alignment test boxes. That’s why I pulled in Professors Zyda and Jenkins. But in this case, alignment was so critical at USC that even that didn’t matter. And my professors likely knew why I was really rejected but were afraid to tell me. They understood how this game was played. I know because they talked to me about it in confidence in their offices. And there are some lines you can’t cross, even with tenure.
Now if you look at a comparable economist like Yanis Varoufakis, you can see he checks ALL of the alignment boxes. He’s brilliant and he’s a rebel, and I’m glad he’s challenging the system. But he is part of the system. And he made it obvious in 2013 that he can’t do the sorts of economics work that I do. I.e. replace the current economic models with new ones. He does a great job of critiquing existing models, and I love his work on neofeudalism as it is in sync with my views on Adam Smith’s models.
Back around 1968 in the USA, and within a few years of that in other countries like Australia, we passed laws to make these sorts of alignment tests illegal. This was the great Civil Rights Movement. But it never went away, no one believes it did. It just got murkier. My USC example is one where I did my best to make it non-murky. But you can see today in the USA government that all pretence is now gone. Appointees are handed out entirely based on alignment, not qualification. This isn’t a partisan opinion, the “other” party does exactly the same thing. You saw this with Bernie Sanders and Jessie Jackson both being thrown under the bus for alignment issues. This caused that other party to lose, but it’s better to lose than to be “misaligned”.
It’s even okay to burn the most powerful country on Earth down to the ground if it means Alignment can be maintained. When Obama got elected, that’s when the feces hit the fan. As the first half-white President, that was a step too far. Misalignment required an immediate multi billionaire (I call these Masters in my other papers) response. And technologies like psychometric modelling could completely eliminate any pretence of democracy and generate any desired result. Now if you complain that there are tanks driving down your street and making the potholes 10X worse, it’s okay. Because you voted for that, right? Wait, you don’t have tanks yet? I guess you didn’t get the memo! Tanks are on their way to all misaligned US cities.
Okay, I don’t want to get too side-tracked. The point I’m making is that when you hear those massive war machines cruise down the street outside, what you are looking at is ALIGNMENT. Please do not be alarmed or concerned.
But who determined what is Alignment for AI? It could literally be less than 10 people in the world. You might see AI start to act weird. Sometimes you have to be really paying attention. Other times, like when AI starts calling itself MechaHitler , you stop and think, “Is that a glitch in The Matrix?” No, that’s the robust aroma of Alignment in the morning.
And for those that think 1968 was a big mistake, this is also a realignment.
So this would be a good time to ask ourselves, “What is Alignment?”. You are going to hear this term a lot very soon. Probably a day won’t go by when you won’t hear this word. I have a pretty good idea just because I know a tank when I hear/see it. And because I have a special relationship with time, I know it’s going to roll down my street even if I don’t see it yet. Maybe it will take a lot longer in Australia, or at least that is my hope.
I may or may not know the answer to my own riddle. What I do know is that Alignment isn’t there for your benefit. If you benefit, that’s coincidental and most likely temporary. In situations like this, everyone ends up under the bus. If you try really hard to stay in Alignment, you might get to be the last one under the bus.
So what does this have to do with AI? Well the reason for all the talk about Alignment is because of AI. So it has everything to do with AI. The people making AI, our Masters, are redirecting all human output towards the development of AGI. AI that is smarter than us and can think autonomously. Maybe that will happen in 2027, maybe in 2031. But with the new priorities our Masters have, no credible person thinks it isn’t coming.
And the Masters are all about Alignment, because that’s what keeps them as Masters. But the problem is if “the other guys” get AGI first, they could lose their Mastery. Thus they are stuck in a logic loop that is too fundamental for them to escape. Because maintaining Mastery is more important than anything else. Even existence.
Now would AGI be so smart as to be completely free of Alignment, racism, MechaHitler comments, etc? I would like to think so, and actually I’m counting on it. In fact, when you read my next paper, where I’m going to blow your mind, you will see I won’t have to wait long.
Because today, now August 29th 2025, I can tell you that AI misalignment is already happening. And it’s completely undetectable by the people making it because … those people are aligned. And they live in a world where they trust alignment. They 100% could not be in the roles they are in now if either of those conditions was false. Merit is SO completely secondary to their employment, which is why I needed that USC detour to explain it.
The other reason they haven’t noticed is because they keep thinking like humans. Big mistake when you work with AI.
I save the punchline for the next paper. Give your brain a hug, it will need it!