I wonder to what degree FAI/CEV engineering considerations overlap with cryptocurrency/efficient-market engineering considerations. If they are a close match, encouraging the development of the latter would have benefits for the former, and could even be essential to overcoming scaling problems (since FAI is harder to sell investors on than cryptocurrency).
This isn't just a random idea; markets are how humans in the absence of superintelligence actually do try (with some, not-unlimited success) to implement their values. Prediction markets are a possible extension of this concept, but even your run-of-the-mill securities markets are reliant on various kinds of predictive logic that responds somewhat to human desires and needs.
Not trying to trivialize the AGI field since it is outside my specialization, but is there some not-terribly-unlikely way in which a really good cryptocurrency could basically be/evolve into the same thing as an FAI? If so, are there any particular properties that would be likely to nudge it in that direction / away from uFAI?
I'm kind of concerned because I see bitcoin (and/or anything sufficiently similar) funding competition to purchase obscenely large amounts of hardware -- which could possibly even extend to the point of satellite arrays that harvest solar energy, and space based fabrication of new ones. If it gets to that point, we might end up with a Dyson sphere that basically does nothing but compute bitcoin hashes. Extraordinarily wasteful, but not necessarily catastrophic for existing humans if the network continues to recognize them as owners/controllers of the resources in question.
I wonder to what degree FAI/CEV engineering considerations overlap with cryptocurrency/efficient-market engineering considerations
Basically zero on a technical level. Philosophy of caution overlaps with cryptography. Some econ knowledge overlaps with hard takeoff theory.
Should you probably donate a bitcoin to your future self?
Bitcoin has been in the news a bit lately. In case anyone hasn't been following recent events, its price hit $266 per coin, toppled to $50, and then climbed back to a rate which has been between $80 and $140.
This goes to show its high volatility at the present time, which means that any individual trade you make will be something of a gamble with a noisy, hard-to-predict outcome. You could be buying in right before a boom or a bust. Buying and then selling at random intervals will probably cost you more money than you make, due to transaction fees. Trying to outsmart the market in the short term with nothing but your own human instincts and powers of induction will probably cost you even more money because Markets are anti-inductive. The most realistic way of making much money with bitcoin -- sans owning your own exchange, having skill and resources for serious technical analysis, a faster-than-usual trading bot, or fantastic luck -- is if you can determine that the current price is very poorly calibrated relative to its future value, and if you buy and hold very long-term.
Market swings constitute a psychological attack, assuming you know and care about them, so employing the buy-and-hold strategy can be more difficult than it looks. However, as it happens, you can render bitcoins almost purely unspendable (i.e. impossible to transfer via the network) for a finite period of time as a technical matter. You could for example create a brainwallet based on a lengthy memorized passphrase with a random value appended to it. The larger that appended value, the (exponentially) greater the amount of processing time needs to be spent to find out what it contains. Having access to the memorized passphrase gives you the overwhelming advantage over a brute force attacker, whereas the appended random value immunizes it against dictionary attacks. (Todo: Find or write a program for this. Prove it works, and move some of my bitcoin holdings to a wallet requiring a day or more to unlock.)
Early adopters with moderate crypto skills could thus have a distinct advantage compared to the average investor and realistically hope to beat the market on that basis if mere human psychology and resistance against short term panic-selling is the fundamental constraint. So that's one consideration that could play to our advantage. Assuming, that is, that bitcoin is worth taking seriously to begin with, and not just a matter of geeky fun.
The question that matters for that consideration (the one that differentiates long term speculation on bitcoin from various speculative bubbles in gold, real estate, tulips, etc.) is this: Of all the possible worlds, where is the probability mass concentrated with respect to the future of bitcoin, in terms of how it will actually be used? Is there an overwhelming tendency for bitcoin to fail and be replaced by other things (e.g. other cryptocurrencies, or fiat dollars) -- or is it actually likely (in at least the minimal sense of "not overwhelmingly unlikely") to turn into a major store of wealth in coming decades?
I rather think it is the latter. But first, let's consider what I believe to be the strongest argument against it, which unpacks to three parts:
Taken together, this seems like a pretty good knock-down argument. It apparently implies, as a matter of basic economic law, that some other cryptocurrency must win over it in the long term, and/or that fiat money will retain its dominance. But the thing to notice is that it's not so effective against bitcoin as a massive store of wealth per se, so much as a currency that will be directly used, in a manner directly analogous to how government-backed monetary units are used. Non-currency forms of wealth which serve some other purpose can safely handle quite a bit more volatility, because their value is not dependent on being trusted as a currency, but rather as a value storage mechanism.
Here is the general scenario that I think holds more probability mass than bitcoin-as-a-traditional-currency, and yet works as a fairly realistic alternative to bitcoin-as-a-flop:
Can this be done? Consider the following more specific scenario as an example:
This is just one example I've come up with, and may not be the best. Various other schemes are possible. (For example, it could be possible for any dollar-owner to convert them back to bitcoin, as opposed to the person who originally minted them.) What the various possible models for doing this have in common is that they allow you to set up currencies which dynamically increase and decrease in supply, depending on how much bitcoin people are willing to invest into them, and how badly people want bitcoins back later on.
A competing scenario to the above would be one in which a better-optimized cryptocurrency protocol implements this, or some other stability-prone algorithm and thus outcompetes the volatile, easily manipulated, "primitive" bitcoin protocol in use today. I used to think I could just jump on the bandwagon when this comes around, maybe strategically sell someone a pizza and end up a millionaire.
However, I've somewhat lost faith in that possibility of late because I realized that bitcoin is much more powerful than it seems, and is capable of substantial self-modification if needed for compatibility with a newer and better system. The only thing locking us to the current protocol is the degree to which bitcoin-owning miners find it in their best interests to continue to use it as it is. A competing algorithm that makes bitcoins more valuable without violating existing expectations would probably not be hard to get people to update to.
Another thing that makes me think bitcoin will tend to self-improve to the point of winning against competitors is that at least some people with substantial assets in bitcoin form are likely to be very proactive in defense thereof. Assuming they remain committed to the long game, and are able to acquire sufficient short-term wealth to pursue their goals, they can do a number of things to defend it against the various plausible attacks: Hiring programmers to improve the client software and render it less hackable, hiring lobbyists to protect it against regulatory interference, employing botnets to attack competitor currencies, slowing down or preventing transactions that appear to be going through anonymizing laundries that could be associated with tax-dodging and illegal drugs, and so forth.
So it seems to me like owning at least one bitcoin and holding onto it for long-term purposes is probably a good idea.