Creators, Computers, and The Concept of Reality
In case you don't have enough to be terrified about already, let's have a glance at the current thing.
Warning: this may be a difficult post for some of you to read. Change can often be scary and I’ve become somewhat convinced things are going to change faster than we may even be able to comprehend. There is always opportunity with change. But there are also people who aren’t going to like that change as well. It has taken me several days to write this and I even contemplated not sharing it at all. Clearly, I have decided to send it. What I don’t want this to be is a trigger for nihilism. I still do firmly believe humanity has a bright future.
I have a friend from elementary school who is one of the more unique people I’ve ever encountered. No doubt hugely eccentric, but I would argue possibly the most intelligent person in our group - and that’s obviously saying something because you’ve all likely gathered how brilliant I am. Jest. Two of my closest friends in high school had 35 ACT scores. I think my best score was a 26. Clearly, I was the village idiot.
My friend who I referred to at the beginning of this was not one of the 35’s though I’m sure he was closer to 35 than I was. When we were in high school, he was still in my circle of friends and we’d occasionally talk about what our future plans were. It’s the simple question often asked of kids young and old…
What do you want to be when you grow up?
The rest of us, myself included, sort of saw his answer as ridiculous. However, with the benefit of 20 years of hindsight, he might not have actually been too crazy after all. His answer:
Philosopher.
Cool story, bro. Now what are you really going to do with your life?
I’m not sure I’ll ever know if my buddy was joking about being a philosopher at the time or not - he didn’t seem to remember it when I brought it up recently - but if there is anyone in my personal life who I’d want to drink whiskey with while talking about the meaning of life, he’d probably be near the top of my list. Why? Because of his philosophical approach to certain things. Inquisitive. Funnier than most. Pretty good at reading people. More intelligent than his attendance record would probably suggest. But mainly just unapologetically weird. He’s philosopher in my book!
Since then, I’ve sort of put myself in his shoes and tried to determine what he was really saying at that time and I keep coming back to this; he just wanted to think for a living. And in many ways, that’s exactly what he ended up doing. That’s what many of us ended up doing.
Now here’s the scary part; I believe human thought is at a bit of crossroads. We’ve become so accustomed to technological advances making our lives easier that many of us probably can’t even comprehend a world where information isn’t right at our fingertips. Ask Google GOOG 0.00%↑ a question, get an immediate answer - sort of. Want to know a specific year something occurred, Wikipedia probably has you covered.
This expectation of near-immediate information is going to get even more intense when the robots start thinking for us to an even larger degree. And this future could be a lot closer than we might appreciate.
Transactional vs Meaningful
I would argue that most of the really great publications on Substack that I read do actually skew more philosophical in nature. But even if what they do doesn’t qualify to be what an academic would consider “contemporary philosophy,” at a base level the writers are just solid free thinkers. I do try to keep Heretic Speculator closer to philosophy than analysis, but analysis and philosophy often intersect. The analysis leads the philosophy. Or is it the other way around?
Hmmm…
Imagine, for a moment, you operate a research service that focuses on analyzing equities, cryptocurrencies, and closed end funds. To be sure, this service offers deeper dives that intend to be longer lasting and more impactful. But there is also a fair amount of lower shelf-life content - this is the more transactional content. It’s not bad. It’s just probably not as meaningful as what you try to accomplish with longer form material.
Now imagine that simultaneously, you also have a blog that is a bit more ‘fringe’ than what might be viewed as actionable investment content - with this blog , you try to provide posts that are more thought-provoking and less ‘transactional’ than some of the other things you produce. Imagine you spend a lot of time on that blog insinuating how fiat currency is a Ponzi scheme.
Narrator: He’s been talking about himself these last two paragraphs.
Now imagine you’re you - shouldn’t be too difficult, right? And even though you may have some general interest in content as a product, if there is a way to get a similar experience in a more cost-effective structure, you’d be silly not to at least consider it.
If you’ve already figured out where I’m going with this, bravo. If not, let me spell it out. GPT-4 is the latest iteration of OpenAI’s generative response tool. This tool, and several others like it that are owned by companies like Microsoft MSFT 0.00%↑ and Google, have been making all sorts of waves in the “creator economy.” For a while, I’ve felt the transactional material is replaceable by the robots but the more meaningful material probably isn't; but I’m actually not as sure about that anymore.
Artists are becoming very upset about it. I’ve seen other writers begin to connect some dots. The possibility of significant disruption to the creator economy by AI is very real and very powerful. Financial analysts aren't safe. And professional bloggers might not be safe either. In BlockChain Reaction, I spend time dissecting the 'tokenomics' or "token economics" of various crypto protocols. On Heretic Speculator, I spend time bitching about lamenting the dollar system. This is what AI said about the “tokeneconomics” of the US Dollar:
You can read if for yourselves if you wish. Personally, I have several takeaways, here are just three:
It’s an incredible commentary about the dollar and what gives it value.
Notably, it returns information about USDC specifically rather than the dollar itself. Likely because the word “crypto” was used in the input query.
Though it’s not perfect, it’s actually pretty good analysis.
On point 3, it’s not all that different from what I do both here and in BCR - it’s actually a strange hybrid of the two. Which means the abilities of these generative tools to mimic human creation is only going to get better. And this is a very big headwind for a lot of industries, including mine. I really don’t want this to come of as hyperbolic and I recognize that these AI bots can be ‘calibrated’ in a way that is clearly influenced by human bias; for example, we’ve already observed the same robot will write a poem gushing about Biden but not Trump. That kind of “AI” isn’t really what I’m talking about.
In a given week, between Seeking Alpha, BlockChain Reaction, and Heretic Speculator, I’d say I generally write anywhere between 6,000 to 8,000 words. Sometimes more. But generally if I try to go too far above that it comes with a quality cost. Time spent not writing is usually spent in the weeds. Toying around with data sets looking for some sort of signal or edge that I can personally act on and then monetize through sharing with others.
But the larger takeaway is pretty simple; in the future, the work I do on a day to day basis is at risk of being disrupted by artificial intelligence. And that’s okay provided I use AI to my advantage rather than let it completely destroy what I’ve been building. To be more clear, I’m talking about process not product. There is a difference. I believe things like GPT-4 can be both a headwind and a tailwind all at the same time. It will be scary to some and an opportunity for others. Each content platform will likely create internal AI policies.
In case you’re wondering, no, I do not use these sorts of things to write anything. Everything you see is me, not a machine. But there will come a time when the machine is better at the tokenomics analysis than I can possibly be. Hell, it might be already. Don’t tell anyone, deal?
The Creative Economy
What about the more creative or “free thought” style content? Surely the blog component of what I do is a bit more safe, right? Maybe. But who really knows? AI is already disrupting things like art. AI can already write poems and even jokes in the style of specific comedians. AI can master language, cadence, and imagery. Imagine what happens when Prince, Elvis, or even Frank Sinatra return from the grave to drop a new song. Crazy? No way:
These tools are far from perfect. But they’re going to get better. Imagine supporting a musician for decades only to learn it was never a real person. I’ve written about Nas several times in the past. He’s my favorite rapper, by far. I’ve always found his music to be thought provoking - dare I say… philosophical. Several years ago he recorded a track called “Death Anniversary.” This is the first sentence from his opening verse of that cut:
What if I told you that AZ didn’t exist, and I put it there, played it like a ventriloquist?
-Nas
AZ is another NY-based rapper who Nas frequently worked with in the 90’s - he’s most definitely a real person. But around the same time of “Death Anniversary” Nas also released the double album Street’s Disciple. On that album he recorded a couple tracks as a different character named “Scarlett” and used technology to change his voice.
Since then, he’s become a successful tech investor. My point is, if there is anyone in music now who might try using AI for creative purposes as opposed to being threatened by it, he’s probably near the top of the list. And wouldn’t you know it, Nas has released 4 solo albums in three years between 2020-2022. Which is a rate of content production he had never accomplished between 1994-2019 when he was “young.”
Now to be clear, I’m not saying Nas has used AI to write or record his recent albums - lockdowns and control of masters are likely the bigger catalysts that drove the surge in content. But how could anyone doubt it if he ultimately says he did use AI down the line? It would be a simple example of Nas using the tools at his disposal. He’s a sharp guy who we know has explored “fake out” content in the past. Creative people explore creation with the tools they have no matter what those tools are.
This is remarkable stuff. It’s powerful. It’s fascinating. It’s incredibly dangerous in the wrong hands. But it’s also going to allow people to explore things they wouldn’t have been able to explore before. My hope is that transactional content will matter a lot less and meaningful content will matter a lot more. When something is so easily replicated by a computer, the value of a human performing the same function diminishes. There are humans who are capable of multiplying large numbers in their heads. But most of us probably wouldn’t consider paying them to do math for us when we can just use a calculator.
The Dangers: AI As A Weapon
It’s not difficult to see how this can be abused. Because if you can use AI to create a song where Kanye West praises Jerry Seinfeld, you can use AI to create a song where Kanye West lobs epithets at Jerry Seinfeld. We are right at the beginning of what could be a troubling era of believable deep fakes. And this will be very damaging to trust and truth - somehow, there’s always another floor below.
Imagine a world where you have a group of friends that you only know through social media and none of them are real. Imagine a world where you see devastation or calamity online and none of it actually happened. To be clear, this isn’t me getting on a soap box and saying “we need to do something to stop AI.” There is probably nothing we can do about it… except unplug it when we have to.
The Opportunities: AI As A Tool
Here’s my hope; AI will probably simplify a lot of the really low-value content creation. And ultimately, it’s probably not a bad thing if that model is disrupted. But people need to still be able to create with or without AI.
I think about how I would personally use AI if it advanced to the degree that I think it will. I have an idea for an app that I believe would be really cool - at the risk of getting Zuckerberg’d, I won’t share it publicly. I’m not technically savvy enough to be able to build this application but I may not ultimately need to be. If AI can write code better than humans down the line, I could theoretically put the idea into a code generator and have AI build the app for me.
Keeping AI In Check
If I can be a bit overt for a moment, you’re kidding yourself if you don’t think cryptography can solve some of the deep fake problems that AI is sure to bring us. If we think about what this blog has mostly been about over the last two years or so, it’s truth and trust. We’ve lost trust in what truth is because truth has been intentionally obfuscated by the “experts” who claim to be the arbiters of what truth is.
This is why cryptography is important but it is absolutely crucial that it exists in a distributed, decentralized form. This is where we get into the Digital ID system and how dangerous something like that can be if we don’t proactively choose the correct approach. And I do mean choose the correct approach because if we don’t choose the maniacs at Davos will push something through anyway and trap the individuals and small businesses that weren’t proactive. Shifting back from digital to analog can certainly work in a localized environment, but that really can’t scale.
Maybe the Amish are on to something? Nah, I like being borderless.
Digital IDs
His fans should know what Kanye’s public key is. They should never know his private key - and neither should the state nor his custodians. He, and he alone, should be the sole controller of what his public key reveals about him. And to be clear, this goes well beyond finance. I’ve already smashed the drum enough on privacy coins like Zcash at this point. You either get it or you don’t. But the application of money is not what I’m talking about.
There are crypto projects, some of which I’ve actually covered in BlockChain Reaction, that specifically address authenticity. Kanye West, or any artist, should be able prove cryptographically if something was made by him or if something was not made by him.
We can use zero-knowledge cryptography to prove we are who we say we are without sacrificing all of our privacy. We can do this in an non-intrusive way and with distributed ledger technology that no central entity can control or manipulate. Freedom seeking individuals and small business owners that don’t want to be stuck in a digital dystopia would be wise to think on these themes and minimize their reliance on ‘the grid.’
That’s the dream at least. As I see it, the only other viable alternative will be simply unplugging ourselves from the Matrix and just assuming everything we see on a screen is a fake out or a psyop. That’s not the world I want to live in though. I’d rather be a doomer optimist than a doomer.
What if I told you Gato Malo didn’t exist, and I put it there, played it like a ventriloquist?
-Pepe Silvia