Controversial uses of prediction markets, should we discuss ethical concerns?

[Akado]

I was thinking on doing a new thread but thankfully found this one.

My question,

Could Prediction Markets potentially make attacks on cryptocurrencies cheaper?

Take for example Bitcoin, which uses POW to secure the network. Currently someone would need to spend thousands or millions of dollars per day to maintaim a 51% attack, that is costly and hard to do.
However with prediction markets, someone with enough money could incentive others to perform other kind of attacks by themselves. Imagine someone creates an event "Exchange X, which has the most BTC volume, will be hacked and go down by the end of the year, 31/12/2015 23:58", then places a big enough amount of money behind their prediction as to get others to get in the scheme and end up being rewarded. We all know how easy exchanges can be exploited. A coordinated attack could place under attack major exchanges as a way to disturb the markets. Ok, this would only affect the price you say, but what about mining pool's? One could condition mining pools as to get one to have >51% mining power.

What do you think about this?

Meanwhile in a parallel universe where all BTS users are scumbags: everyone gets behind a prediction where all major altcoin exchanges get hacked. Someone will eventually find it worthy and decide to do it. Altcoin exchanges go down. Coordinate this as a mean to let all exchanges who run on top of BitShares (like CCEDK) unharmed. BitShares now holds all the altcoin volume. Profit.

Whoever decides to do this - and speculating a lot i know - could just start an exchange/mining pool's war, messing up the market. But hey, this is  free market after all

[luckybit]

At the end of the day it all comes down to marketing.  A market is nothing if there is no one trading it in and that means people must advertise the market.

Wallet hosting providers will get to filter which markets they allow users to trade in on their website.   This means that no ethically controversial prediction markets will be tradable on the most popular, easiest to use, mainstream hosted wallets.   This means that people will have to download full clients.   Full clients will default to only showing the most heavily traded / trending markets which means there would be no organic discovery. 

At the end of the day, these morally controversial markets will be relegated to the dark corners of the blockchain where no one sees them and the trading fees generated by them are not enough to cover the asset registration fees or advertising costs.

The "decider" in private prediction markets would likely be legally liable and therefore few would dare to create such a market.

A very thoughtful and considerate response. This is the kind of leadership the community needs to see.

The solution you propose is also very feasible, it strikes a good balance. What does everyone else think?

[bytemaster]

At the end of the day it all comes down to marketing.  A market is nothing if there is no one trading it in and that means people must advertise the market.

Wallet hosting providers will get to filter which markets they allow users to trade in on their website.   This means that no ethically controversial prediction markets will be tradable on the most popular, easiest to use, mainstream hosted wallets.   This means that people will have to download full clients.   Full clients will default to only showing the most heavily traded / trending markets which means there would be no organic discovery. 

At the end of the day, these morally controversial markets will be relegated to the dark corners of the blockchain where no one sees them and the trading fees generated by them are not enough to cover the asset registration fees or advertising costs.

The "decider" in private prediction markets would likely be legally liable and therefore few would dare to create such a market.

[xiahui135]

I don't think prediction market is important and need to exist.

Consider what we have, is it the best thing we can do ?

[betax]

Any system will eventually provide auto censorship, due to ethical issues. In the original op we see somebody asking what cashier will be fired next, what are the consequences of this?

If a person working for the company is responsible for this, it will have a negative repercussion on the company, as nobody will think twice to work there and many people will not want to shop there.

Prediction markets on illegal issues or bullying will be quickly put down. Like everything, if is in the open there are always means to investigate and find, much better than underground. Bitshares is not anonymous after all.

Sadly there is always going to be corruption and people will try to manipulate the markets, but this is what we are trying to fight by decentralising.

[Troglodactyl]

I don't think this is a problem that needs to be solved.  Prediction markets can have many use cases including as a crowd funding system, but that doesn't mean we need a system for censoring them.

[luckybit]

in terms of ethic, here's how Augur handles it.

They're selling REP tokens to end users, who in return will have voting power on the outcome of an event (check their FAQ on augur.net, can't link it at the moment).

however, there will be an option to vote a prediction market out due to ethical concerns.

This is the sort of solution which Bitshares could consider implementing. In the case where a prediction market arises on the platform which threatens the integrity of the platform or safety of the members of the community. But I don't know exactly how to do that.

Like for example you could connect it to reputation. So that there are official prediction markets where a person stakes their reputation on the line, and unofficial.

[luckybit]

Efficient, liquid markets only form when there is enough public information on the underlying. It is less likely that a market would form around the unemployment or incarceration prospects of a little know individual as opposed to a celebrity. Prediction markets won't really affect the average joe, but they could effect celebrities.

You're talking about anonymously. Bitshares users are not anonymous. Anyone on this forum or involved with Bitshares to the degree we are, are all known. We all probably have files in some government database if you want to be paranoid about it.

I think when it's involving celebrities, you're talking about the most powerful people on earth, with plenty of money? And when it's assassination then it's like declaring war. I don't think it would be something which will be very popularly because if you think about it the people who have the most money in society can afford the body guards, they know the politicians or maybe even the President, and there isn't any privacy anymore.

On the other hand on the small scale, such as a boss using prediction markets against his employees, that kind of situation is what I would expect. Mainly because while there might be the court case in the United States showing it not to be legally wise, in another country where there hasn't been any court ruling the situation is different, the risk profile is different,  and unlike assassination which is universally off limits, there are cultural differences which would make certain kinds of prediction markets more likely.

Additionally, I think that assassination markets are actually more likely than the other examples you provided. Betting on the death of celebrities is a far more common thing than you would assume.  Assassination markets could pose a really big problem in the future if permissionless blockchains take off. On Bitshares, delegates still need to approve of the creation of a given market.

Then perhaps we should discuss now what the approval process should be, otherwise these prediction markets could be used to destroy Bitshares. I know with Augur there are some questions which can be really bad and there are ways to reject the question as invalid.

I don't think that the market would be established as you have framed it. A better market would be for the probability that some well known figure is not simply indicted but rather found guilty and sentenced to prison time.

Either way you could create prediction markets around that, and it could easily become something terrible.

In your example if a law enforcement officer were to arrest a well known figure without cause, people would know that there was a market and incentive for that officer to unjustly arrest the person and that officer would have to face whatever the associated consequences were.

They could come up with a cause prior to arrest. Tax evasion for instance or perhaps some obscure law was broken. The point is that it can be turned against us in the long term.

I'm not saying that benefits could not outweigh the costs, I'm just saying that there are deterrents for officers to unjustly arrest people even given an arrest market.

What deterrents? Do you really think if police really wanted to arrest and convict someone that it couldn't be arranged to happen?

Also most participants would enter into markets that could not easily be gamed, which is why I suggested that the market would probably pertain to conviction and not indictment.

And you think convictions can't be gamed? https://en.wikipedia.org/wiki/Kids_for_cash_scandal

I don't think Bitshares will be agnostic, at least not with the current delegates. Bitshares will represent the userbase. If the userbase gets really big, as in millions of people, then perhaps at that point the prediction market could be turned against us, which is why it's important to consider ethical implications.

The people currently involved with Bitshares are ethically oriented, and as long as Bitshares has self governing mechanisms it can possibly avoid most of the negative side effects and achieve the positive benefits. A complete hands off approach on the other hand, or an amoral approach, might not work out so well in the long term.

Bitshares has to in my opinion attract the most ethical people to make up it's userbase, the most responsible people, people who aren't reckless. Not every platform is going to operate on a set of principles to care to even weigh the positives and negatives. Some platforms will be completely anonymous (or at least the users will believe they are), and those platforms are where you'll likely see the unpredictable consequences.

[fav]

in terms of ethic, here's how Augur handles it.

They're selling REP tokens to end users, who in return will have voting power on the outcome of an event (check their FAQ on augur.net, can't link it at the moment).

however, there will be an option to vote a prediction market out due to ethical concerns.

[Empirical1.2]

I think the market for something like assassination PM's will be niche & so morally repugnant to the vast majority of BTS customers that they won't be allowed because the net value loss from losing customers who are against it would be too great.

It's similar to front-running. It's extremely unpopular & so given the choice between two similar blockchains, where one allows blatant and excessive front-running and one that limits it. The majority of customers will choose the latter making the former the loser.

So basically any activity the vast majority strongly disagree with, enough to switch blockchains, will probably not be present/severely restricted on the major/market leader blockchains. 

[seocuenta]

I think it is Winner Winner, they are catching many people's attention and that clearly means it will be much more powerful!

[clout]

This is all true. The cashier fired is one example, the arrest example is another, and the assassination example is another. The assassination example is very risky, and not as likely to be a common occurrence, but the less risky examples may become common.

Efficient, liquid markets only form when there is enough public information on the underlying. It is less likely that a market would form around the unemployment or incarceration prospects of a little know individual as opposed to a celebrity. Prediction markets won't really affect the average joe, but they could effect celebrities.

Additionally, I think that assassination markets are actually more likely than the other examples you provided. Betting on the death of celebrities is a far more common thing than you would assume.  Assassination markets could pose a really big problem in the future if permissionless blockchains take off. On Bitshares, delegates still need to approve of the creation of a given market.

There could be arrest prediction markets, and the people in law enforcement might have the incentive to bet in those kinds of markets considering they have what seems like free reign to decide who gets arrested and for what reasons.

I don't think that the market would be established as you have framed it. A better market would be for the probability that some well known figure is not simply indicted but rather found guilty and sentenced to prison time.

In your example if a law enforcement officer were to arrest a well known figure without cause, people would know that there was a market and incentive for that officer to unjustly arrest the person and that officer would have to face whatever the associated consequences were. I'm not saying that benefits could not outweigh the costs, I'm just saying that there are deterrents for officers to unjustly arrest people even given an arrest market. Also most participants would enter into markets that could not easily be gamed, which is why I suggested that the market would probably pertain to conviction and not indictment.

We should at least understand the can of worms prediction markets are opening up, and the real risks they could bring. I don't think Bitshares is irresponsible about these things but there are other communities which don't care to be responsible.

At the end of the day, Bitshares is building an agnostic technology, what people do with it is up to them.

[mint chocolate chip]

Is Augur crowd sale a good investment you think?

That is a question to ask on the Ethereum forums. Shouldn't we be talking about the Bitshares 2.0 prediction market?

That is a potential future feature I believe, don't think it is part of the 2.0 release so I would think asking the community's opinion on a competing prediction market when you mention it in the OP is alright.

It is part of 2.0 from my understanding, although it might not be a sophisticated as Augur.

Hope it is, sure will be fun https://bitsharestalk.org/index.php?topic=17777.0

[luckybit]

I don't see how there is anything ethically wrong with the example in OP. There is nothing wrong with publicly validating information.

The problem is when a market incentives unethical courses of action (i.e. trying to get a cashier fired or assassinating someone to profit from these markets). Whats most troubling is that this phenomenon would most adversely impact the people that are least likely to be fired or assassinated because the payout would be that much larger.

This is all true. The cashier fired is one example, the arrest example is another, and the assassination example is another. The assassination example is very risky, and not as likely to be a common occurrence, but the less risky examples may become common. If people are fired they pretty much are in a state of fear and in the example the boss sent that out so he was in the position to decide who gets fired anyway and wanted to instill perpetual fear in his underlings.

But there is more. We saw what happened with the corrupt DEA agents a while back and how the police can be corrupt to that extent. There could be arrest prediction markets, and the people in law enforcement might have the incentive to bet in those kinds of markets considering they have what seems like free reign to decide who gets arrested and for what reasons.

We should at least understand the can of worms prediction markets are opening up, and the real risks they could bring. I don't think Bitshares is irresponsible about these things but there are other communities which don't care to be responsible.

http://usatoday30.usatoday.com/news/nation/story/2011-10-03/boss-firing-contest-unemployment-benefits-ruling/50647688/1

http://blogs.findlaw.com/free_enterprise/2011/10/10-if-you-guess-who-will-be-fired-next.html

[luckybit]

As a neutral platform, BitShares will enable people to network and collaborate with others more effectively.  I don't think there's a need to cripple it just because some people may use it for things of which we disapprove in the future.

You might think it's neutral but other people might use your own markets against you.

Keep in mind that even the most dreaded "assassination market" application of prediction markets is just as much an opportunity to crowd fund the defense of the person as it is an opportunity to crowd fund the assassination. 

There is a difference in professionalism. Private security services don't select just anyone off the street, and if you get into the realm of assassination then it becomes warfare. At war it would be obvious that the government which currently has the best weapons and best trained killers would win. So it's not really something we would benefit from, and the tool would more likely be turned against it's creators, sort of like how the atomic bomb didn't benefit the creators.

At its core, this is a fundamental question about the aggregate nature of humanity.  If you think new technologies that increase human productivity are a net negative because they enable people to produce more misery and suffering, this may not be the community for you.  If you think that technology and increased productivity is generally an improvement and leaves people better off, then check your paranoia and have some appreciation for what this community is building for human empowerment.

Never said it's a net negative or net positive, but I think we have to be responsible in our implementation and be willing to discuss all ethical and social implications of any tool we endorse, or help create. If we believe it is good, then we should use it for good, but the truth is that not everyone will care about human life, human empowerment, productivity, or any of that.

I'm not even talking about assassination markets although those markets would obviously be problematic. I'm talking about prediction markets which could create abuse without anyone being killed, because most of the time people can be hurt without anyone being killed.

Example would be the prediction markets around hiring and firing, or prediction markets around who will be arrested next, or prediction markets which are hostile towards Bitshares itself. It's something to debate, because any technology can be abused and turned against it's founders.