Artificial Intelligence, economics and the famous guy in the wheelchair

We are too confident in our ability to understand and control complex phenomena like the economy or politics.

Stephen Hawking, the famous guy in the wheelchair, has a dire prediction for the future of humanity. According to him, we will wipe ourselves out within the next 1 000 years with a probability of 1 (100%)! According to him, the probability that we will do so within any given year is negligible, but that given enough time, it is a certainty. RIP Homo Sapiens.

While a 1000 years is tiny bit longer than I was expecting to be here, which makes me slightly less anxious about Professor Hawking’s dire prediction, the scenarios that might lead to humanity facing some grave challenges fascinate me. One of these grave challenges is the economy, what shape it is in now, and what might happen to it in the future. Another thing is the rise of the smart machines, the artificially smart ones. You thought terminator was just a movie? Think again. 

Water is composed of 2 hydrogen atoms and one oxygen atom. It has some remarkable properties which is essential to life as we know it. For instance, if water did not expand when cooled from 4 degrees Celsius to its freezing point at 0 degrees, ice wouldn’t float and the oceans and life as we know it would not exist. Another peculiar property of water is that it is wet. Hydrogen atoms aren’t wet, and as far as I know oxygen atoms aren’t wet either, which begs the question.

How is it that water is wet if the two components that its made from aren’t wet? I’m not going to try and explain the mechanics of the wetness of water, you have more important things to do, and I don’t know the answer in any case. Except to say that it seems that the “wetness” somehow emerges from the combination of 2 hydrogens and 1 oxygen. 

Like water, the economy is also composed of individual parts. Seven billion-odd of us in fact. What’s truly amazing is that the property of many of us getting what we want when we want it at a price we are willing (grudgingly!) and able (barely!) to pay (what economists call allocative efficiency and effectiveness), emerges from us essentially doing our own thing! Without deliberately planning for economic order to emerge (markets and such) it just does.

What is more is that whenever a powerful person, or group of people (governments), have tried to unduly control the emergent economic properties of efficiency (doing things well) and effectiveness (doing the right things) they have destroyed the very things that they planned on controlling. Think how poor communism has been in allocating resources both from an efficiency perspective (corruption et al.) and effectiveness (ghost cities in China). Consider the disastrous economic management record of most state-owned enterprises. They may give you what you want (plane tickets or electricity) but they often waste resources unnecessarily in doing so (SAA and Eskom).

We do not really know how the wetness of water emerges from the interaction of hydrogen and oxygen. My master’s thesis (in neuro-philosophy) attempts to add a thought or two to our conversation about the wetness of water amongst other things. What perplexes me is that we are so confident that we know and can control an emergent system (the economy) many orders of magnitude more complex than water. 

There is a simple reason why top down control, of any variety, including the capitalist central bank, liberal financial-markets-sort, is bound to fail from time to time. The reason is that any individual actor (you and me), or small group of actors, has the ability to understand what they want and how they are going to get it to a very large degree and most of the time individually, but none of us can do so for the whole system – not the currency market, not the bond market, not the stock market, not any government or agency, nobody bar none.

This means that sometimes the market is going to get it wrong and we will be none the wiser for a shorter or longer period of time. These market failures will then lead to market corrections, recessions, or even depressions depending on the nature of the mistake and the time it takes us to find out that we’ve been wrong. In the capitalist world we have the market as our most reliable indicator or mechanism to find out if something has gone wrong or is perhaps going wrong. I am very afraid because the market mechanism in the second largest economy in the world is like a typical cheapie – looks good, works not so well.

We have been praising the Chinese for decades for the success of their top-down, state-controlled, hybrid form of capitalism. What if the Chinese economic miracle happened despite the top-down meddling of the communist party and their state-owned enterprises (SOEs)? And that one of the only good things they did was to partly(!) allow the individual actors in the economy to do what humans do – be enterprising and self-interestedly cooperating. What if they allowed complete economic freedom, and meddled minimally to try and correct for market failures and to provide public goods? Would the Chinese economy not be many times larger than its tremendous current size?

My fear is that in China, and by implication the rest of the world economy, the mechanism for knowing when a market has made a mistake (of efficiency and effectiveness), takes an awfully long time to work, because the Chinese mechanism is based on the knowledge and incentives of a relatively small group of people, even though they may be particularly smart. In other words, China is in a systemic position to make mistakes of capital allocation for many years, or even decades, before dire emergent economic properties show their ugly recessionary or depressionary heads, because their mechanism for finding out if they have made economic mistakes is too “uninformed” to understand a complex system such as an economy. They may have been the first to invent and control gunpowder, but the firecracker-economy they are operating at the moment is probably going to result in the biggest bang they, and we, have seen this side of the communist revolution.

This article is getting awfully long, and I haven’t even said anything about Artificial Intelligence (AI) yet. Give me 200 words to wrap up. We are too confident in our ability to understand and control complex phenomena like the economy or politics. This is because new properties not present directly in the constituent parts of a complex system emerge from the interaction of these parts. The fact of the matter is that we do not really understand, and hence cannot expect to effectively control, these properties if and when they do emerge.

My greatest fear is that we are building machines from the combined knowledge of many human beings, which means no single human is smart enough to build it or unbuild it by him or herself. It may emerge that these machines are smarter than us, which means they are smarter at building themselves than what we are at building them, which means we may very well have moved one spot down the galactic food chain in such a scenario.

That’s what I’m afraid of. And you should be too.

Was this article by Charl helpful?


You must be signed in and an Insider Gold subscriber to comment.




Instrument Details  

You do not have any portfolios, please create one here.
You do not have an alert portfolio, please create one here.

Follow us:

Search Articles:
Click a Company: