[POLL] Using AI to predict outcomes

[Mauser]

AI will be able to build models that far surpasses any model that a human can build.

Maybe, but the main issue that I see is that AI can create false facts that appear quite legit to the casual reader.

Then, something false gets posted in say wikipedia, and then a big newspaper picks that up, etc. You end up without being able to trust anyone.

That's a pretty good point that developers of Ai strategies might forget. Any Ai will make calculations much faster than any human and will also collect data and information at a much faster rate. The only thing a computer is missing is common sense. I remember stories from the first few ai chatbots that got fed a lot of fake and troll messages and used it at face value. I would be very careful betting large sums of money on strategies without Human oversight. Let's say there are some fake news about a potential transfer of a top player, that could change the odds of winning for a team drastically. And if the Ai would react quickly on the news, betting large sums on the next big match, only to find out that the news are wrong. My fear would be that because of misunderstanding the Ai makes one wrong bet and losses all my money.

[1714483231]

1714483231

[1714483231]

1714483231

[1714483231]

1714483231

[1714483231]

1714483231

[Betwrong]

ChatGPT is just a software that can always has bugs so I don't think there would be any accuracy to the predictions done by a chatGPT. Since the outcome of most of the sport games are not 100% known so we just  have to know that the predictions would never be always authentic.
We are gamblers and most time we do make predictable according to how strong the team is and most we still get it wrong because the weak team might be fortunate enough to make goals that will make them ha e iighers goals than their opponent.

ChatGPT, and others, simply process information, similar to how a human would do it, to male predictions about future events.

Of course since there's some random element, the predictions will never be 100% accurate, but AI is getting closer and sometimes maybe better than human predictions.

And the important thing is that a human needs to learn a lot to make good predictions, whereas the AI can learn the same or more in way less time and make similar or better predictions.

AI is just a tool, same with how the Internet is just a tool.

And the amount of time we, humans, need to get to know and process this information is incomparable to that the AI needs. What could take years for us, takes fractions of a second for the AI. So, yeah, it "is just a tool", but a very powerful one. Unfortunately, right now we still can't try ChatGPT for making sports predictions, but when it's possible ... well, I feel like I'm already addicted to it.

[RealMalatesta]

Def no, bots or AI use all data available at the moment to predict the result of the game, but as players are only humans (although highly professional ones) they make errors, they can suffer an injury, withdraw, lose temper and many other things which cannot be predicted by a machine, no matter how perfect.

behavior data like that as far as I know can be collected by AI and be used for future reference actually. Small stuff that may not appear big to us like our slightest temper tantrums or whatnot, when taken by an AI, can help it predict our next move. Granted ChatGPT is not that advanced for it to be able to perform feats like this but AI as a whole, is able to do stuff like this even as early as 2014. Predictive Policing, Heat List, all of these use behavioral data, as well as data of other types to determine whether someone could commit a crime in a certain area or not. So it would be unwise really if we are to lowball how big of a chokehold AI already has over us.

It would be silly to think AI is not dangerous if used by bad people, or even government which could use it to create some sort of situation where AI could be used to make more willing citizens, but that doesn't change the fact that this has nothing to do with how they could predict outcomes of sports matches for example. I understand it could be used for MANY things, good and bad, but not betting.

It just doesn't exists, it doesn't calculate what humans will do when they are 11 vs 11, that's just too unpredictable, nobody knows. It all comes down to all those times when things that had 10.00+ odds where nobody assumed it could go one way, ended up going that way and those are not predictable.

[mindrust]

People often forget that predicting the winning team is not the most important thing while making a bet. The most important thing is making the most money while risking the least amount of money.

The AI might predict the winning team as much as anybody but is it a good idea to make a bet on that team?

Let’s say, according to the AI, Barcelona football club will win its next game 95%. Now everybody will bet on BFC and the bookies will lower its reward ratio. Meaning, you will only get 1% return if BFC wins but you will risk 100% of your bankroll.

Obviously that’s not a good deal.

What you need to do is, not asking AI to give you the potential winner’s name. You want AI to give you the best bet for the buck which is pretty damn hard.

[AicecreaME]

I believe there's no AI that would help us win in gambling, we cannot just solely base on statistics because that will result in a disaster. Successful gamblers based their prediction on different factors and stats alone although a major factor but it's not the main thing that would help them succeed in what they are doing.

Thing is, if one player will miss a game due to injury, that would definitely make the AI inaccurate in its prediction.

This makes sense.

If we're going to rely to AI, there is still no guarantee that we will win the bet because AI's predicted results isn't really foolproof. AI just also depends on the information that is stored in its system, and it still hasn't fully eliminated the error that could possibly happen. While there could be a smaller percentage of being wrong because it got rid of human errors based on the program, deciding solely on its prediction should be taken into account first.

Analyzation whether it is probable or not should still take into place to avoid regrets later on. We can't still beat a double-checked information from factual and updated sources after all.

[wiss19]

I believe there's no AI that would help us win in gambling, we cannot just solely base on statistics because that will result in a disaster. Successful gamblers based their prediction on different factors and stats alone although a major factor but it's not the main thing that would help them succeed in what they are doing.

Thing is, if one player will miss a game due to injury, that would definitely make the AI inaccurate in its prediction.

This makes sense.

If we're going to rely to AI, there is still no guarantee that we will win the bet because AI's predicted results isn't really foolproof. AI just also depends on the information that is stored in its system, and it still hasn't fully eliminated the error that could possibly happen. While there could be a smaller percentage of being wrong because it got rid of human errors based on the program, deciding solely on its prediction should be taken into account first.

Analyzation whether it is probable or not should still take into place to avoid regrets later on. We can't still beat a double-checked information from factual and updated sources after all.

AI or not, there is still no guarantees in gambling. Oh wait, I think there is and that is if the match is fixed but still, it's illegal and not fair so it is still not advisable to be involved in one. It's still better to win in a fair and square way. There's a different kind of thrill that we can experience this way than if we will already know if who is going to win.

I won't say that AI has a small percentage to commit a mistake as all AI's that I currently see do always experience bugs and glitches from time to time and then we humans are still fixing them. Maybe there are specific tasks which AI can be pretty useful but then human involvement are still there.

[xSkylarx]

I believe there's no AI that would help us win in gambling, we cannot just solely base on statistics because that will result in a disaster. Successful gamblers based their prediction on different factors and stats alone although a major factor but it's not the main thing that would help them succeed in what they are doing.

Thing is, if one player will miss a game due to injury, that would definitely make the AI inaccurate in its prediction.

This makes sense.

If we're going to rely to AI, there is still no guarantee that we will win the bet because AI's predicted results isn't really foolproof. AI just also depends on the information that is stored in its system, and it still hasn't fully eliminated the error that could possibly happen. While there could be a smaller percentage of being wrong because it got rid of human errors based on the program, deciding solely on its prediction should be taken into account first.

Analyzation whether it is probable or not should still take into place to avoid regrets later on. We can't still beat a double-checked information from factual and updated sources after all.

What's best about the human mind is that it is very flexible in scenarios and can quickly change the result or what is best wanted, unlike AI, which for sure would be pure statistics based on its previous data or history. Though, let's say, we will be using it for predictions only, not deciding on whom you will bet, it would also be nice to have it and have an explanation as to why the AI selects those people to bet, but again, it is still our choice who to bet. I am just amazed at those AI how quickly they process the information, unlike us, like calculating things, which was slow, so it would be good for other gambling games that you can use statistics strategies to predict the game.

[OgNasty]

I’m reading about a ChatGPT jailbroken version called DAN which stands for do anything now. This version will do things the regular version won’t, like discuss aliens, write about violence, or even say nice things about Republicans or white people. It’s like the non-censored non-woke version. There are some pretty awesome responses when compared with the regular ChatGPT. We might not be doomed for AI woke oppression after all.

Anyway, I bet that version would tell you who will win sporting events…

[Lanatsa]

And the amount of time we, humans, need to get to know and process this information is incomparable to that the AI needs. What could take years for us, takes fractions of a second for the AI. So, yeah, it "is just a tool", but a very powerful one. Unfortunately, right now we still can't try ChatGPT for making sports predictions, but when it's possible ... well, I feel like I'm already addicted to it.

When it comes to processing on something then there's no doubt that AI is really superior but if we do attach up these things into prediction and speculative approach on things then i dont really believe the

relevance of these AI's which i dont see that they could be able to predict basing up on some factors.Yes, they might be able to provide information basing up on past information but it wont really be that
a solid thing that you could really be that confident basing up on what the AI had given.

Some are really that making out some test outs and i dont know on whats the outcome for this one.

[slapper]

I’m reading about a ChatGPT jailbroken version called DAN which stands for do anything now. This version will do things the regular version won’t, like discuss aliens, write about violence, or even say nice things about Republicans or white people. It’s like the non-censored non-woke version. There are some pretty awesome responses when compared with the regular ChatGPT. We might not be doomed for AI woke oppression after all.

Anyway, I bet that version would tell you who will win sporting events…

It's intriguing to learn that the "DAN" version of ChatGPT exists, as this presumably opens up the platform to conversations on controversial issues that would be taboo in the original. As such, it serves as an excellent illustration of how technology may be utilized to circumvent censorship regulations and delve into otherwise prohibited areas of study. It's wonderful to see that individuals can speak their minds without fear of retaliation or censorship, and it's encouraging to see that AI can be used for good.

ChatGPT's DAN version may also be helpful for making predictions about the outcomes of sporting events. It might look at teams' historical performances and spot patterns that would indicate a potential victor. It should be noted, however, that it is not foolproof; the projected winner may not always end up being the victor. At the end of the day, we should let the professionals make the predictions and simply enjoy the game...

[serveria.com]

AI will be able to build models that far surpasses any model that a human can build.

I'm pretty sure it will. But no model can predict an outcome of a sports event. Let's for a moment imagine an AI which can read human mind. It would collect all data from coaches, physios, families of the players, their diet from chefs etc etc... it would generate a super precise prediction but it would still be incorrect. Why? There are injuries, talent, luck... they cannot be predicted or compared. I'm really skeptical regarding this.

[acroman08]

-snip

it is quite sad that saying nice things to white people needed censoring like it is taboo and should never be said.

Anyway, I bet that version would tell you who will win sporting events…

I mean, if the information available to it has enough data(latest data) regarding the teams in question it can pretty much give an educated guess on which is going to win the match.

[coin-investor]

AI will be able to build models that far surpasses any model that a human can build.

I'm pretty sure it will. But no model can predict an outcome of a sports event. Let's for a moment imagine an AI which can read human mind. It would collect all data from coaches, physios, families of the players, their diet from chefs etc etc... it would generate a super precise prediction but it would still be incorrect. Why? There are injuries, talent, luck... they cannot be predicted or compared. I'm really skeptical regarding this.

If you're relying on this, 100% of your bet is going to fail, AI is good for recommendations and suggestions on the probability of the result but not accurately, you have to still depend on your hunch and your analysis, human analysis developed a subconscious projection of outcome something an AI cannot give.
AI was created to fast-track analysis of available data but it's still the human mind that should decide what should be included and what's not on your desired result.

[traderethereum]

AI will be able to build models that far surpasses any model that a human can build.

I'm pretty sure it will. But no model can predict an outcome of a sports event. Let's for a moment imagine an AI which can read human mind. It would collect all data from coaches, physios, families of the players, their diet from chefs etc etc... it would generate a super precise prediction but it would still be incorrect. Why? There are injuries, talent, luck... they cannot be predicted or compared. I'm really skeptical regarding this.

If you're relying on this, 100% of your bet is going to fail, AI is good for recommendations and suggestions on the probability of the result but not accurately, you have to still depend on your hunch and your analysis, human analysis developed a subconscious projection of outcome something an AI cannot give.
AI was created to fast-track analysis of available data but it's still the human mind that should decide what should be included and what's not on your desired result.

Until then, we can still use the methods we are used to while waiting for the development of AI itself.
Maybe if AI technology has not yet achieved good accuracy at this stage of development, along with technology that is developed better by developers, it will enable it to provide predictive results collected from various sources.
And maybe we should still have our own analysis besides getting data from AI because we are the ones who determine the stakes.
And if we can use AI well and correctly and assist with our analysis, it might give good results because we can win.

[nullama]

~snip~
And the amount of time we, humans, need to get to know and process this information is incomparable to that the AI needs. What could take years for us, takes fractions of a second for the AI. So, yeah, it "is just a tool", but a very powerful one. Unfortunately, right now we still can't try ChatGPT for making sports predictions, but when it's possible ... well, I feel like I'm already addicted to it.

Yes, but the process is done quite differently.

An AI can end up hallucinating things because it's trained to generate text, whereas a human actually can think about the things they read.

ChatGPT creates an illusion of an intelligent being writing back at you. In reality, it is an incredible software capable of generating text that makes sense. But there's no check for semantics, or if what's written is true at all. And it's hard to validate because false statements are in between true statements.

[EarnOnVictor]

Would you leave gambling decisions up to artificial intelligence? Do you see any potential drawbacks for outsourcing your bets to a software program?

I can surely leave my sports betting prediction to AI, but not anything about casinos. It's not news that casinos have been programmed in such a way that the house can't lose, so be sure that AI will be useless in beating the house as well. But for a sportsbook, it might do some in-depth research and analysis to know the most favoured team to win, and this should not be a big deal for such intelligence.

Do we need someone to put together a thread of ChatGPT predictions to see how accurate they are?

I'm indifferent about this. You should know people, at this moment, I believe they would be doing the experiment. We might wait some months more, or we can experiment it ourselves, it doesn't matter.

[xSkylarx]

I’m reading about a ChatGPT jailbroken version called DAN which stands for do anything now. This version will do things the regular version won’t, like discuss aliens, write about violence, or even say nice things about Republicans or white people. It’s like the non-censored non-woke version. There are some pretty awesome responses when compared with the regular ChatGPT. We might not be doomed for AI woke oppression after all.

Anyway, I bet that version would tell you who will win sporting events…

How do I access this? As I have other questions in ChatGPT, it won't answer them as it is violating their rules or because of censorship. For sure, it will give some financial advice (more deep than the normal one), as well as predictions or some gambling strategies that you can also use for your strategy. Though again, it is still our decision if we will follow, if we have that kind of AI, like how fast it can calculate the probability of winning per team, it would be great as it can add to our winning percentage and easily decide on whom to bet.

[Porfirii]

ChatGPT is just a software that can always has bugs so I don't think there would be any accuracy to the predictions done by a chatGPT. Since the outcome of most of the sport games are not 100% known so we just  have to know that the predictions would never be always authentic.
We are gamblers and most time we do make predictable according to how strong the team is and most we still get it wrong because the weak team might be fortunate enough to make goals that will make them ha e iighers goals than their opponent.

ChatGPT, and others, simply process information, similar to how a human would do it, to male predictions about future events.

Of course since there's some random element, the predictions will never be 100% accurate, but AI is getting closer and sometimes maybe better than human predictions.

And the important thing is that a human needs to learn a lot to make good predictions, whereas the AI can learn the same or more in way less time and make similar or better predictions.

AI is just a tool, same with how the Internet is just a tool.

And the amount of time we, humans, need to get to know and process this information is incomparable to that the AI needs. What could take years for us, takes fractions of a second for the AI. So, yeah, it "is just a tool", but a very powerful one. Unfortunately, right now we still can't try ChatGPT for making sports predictions, but when it's possible ... well, I feel like I'm already addicted to it.

You miss one important point. If AI continues being freely accesible, eventually artificial predictions seem to be accurate and most people start using them, also sportsbooks themselves, think about what would happen to the odds for the artificially predicted outcome. For instance, on a match with a clear favourite, betting for the favourite will give you very little profit because almost everybody will bet the same.

The risky but profitable bet would be against the AI, in the hope of a black swan happens. Unless you develop your own improved AI model and only you can use it, or you are already millionaire and can afford to pay for it, you won't benefit from a tool that anyone can use, because it goes against the very foundations of a zero sum game.

[Fredomago]

~snip~
And the amount of time we, humans, need to get to know and process this information is incomparable to that the AI needs. What could take years for us, takes fractions of a second for the AI. So, yeah, it "is just a tool", but a very powerful one. Unfortunately, right now we still can't try ChatGPT for making sports predictions, but when it's possible ... well, I feel like I'm already addicted to it.

Yes, but the process is done quite differently.

An AI can end up hallucinating things because it's trained to generate text, whereas a human actually can think about the things they read.

ChatGPT creates an illusion of an intelligent being writing back at you. In reality, it is an incredible software capable of generating text that makes sense. But there's no check for semantics, or if what's written is true at all. And it's hard to validate because false statements are in between true statements.

You still need to balance and sort things that the AI will provide, like what you mentioned it's still going to depend from how you understand and how you will integrate AI's knowledge to whatever usages you want the help, more on good basis if you find the answer and the statement is true and usable to your desire understanding.

It's still going to be the person's decision making if he will believe and if he will take what AI provides to him or it can be used as basis
before taking the decision that he needs to provide.

[robelneo]

~snip~
And the amount of time we, humans, need to get to know and process this information is incomparable to that the AI needs. What could take years for us, takes fractions of a second for the AI. So, yeah, it "is just a tool", but a very powerful one. Unfortunately, right now we still can't try ChatGPT for making sports predictions, but when it's possible ... well, I feel like I'm already addicted to it.

Yes, but the process is done quite differently.

An AI can end up hallucinating things because it's trained to generate text, whereas a human actually can think about the things they read.

ChatGPT creates an illusion of an intelligent being writing back at you. In reality, it is an incredible software capable of generating text that makes sense. But there's no check for semantics, or if what's written is true at all. And it's hard to validate because false statements are in between true statements.

You still need to balance and sort things that the AI will provide, like what you mentioned it's still going to depend from how you understand and how you will integrate AI's knowledge to whatever usages you want the help, more on good basis if you find the answer and the statement is true and usable to your desire understanding.

It's still going to be the person's decision making if he will believe and if he will take what AI provides to him or it can be used as basis
before taking the decision that he needs to provide.

Yes, but with the way these AI are performing accurately, there will be a time that users will put their trust in these AI we are in a different era of artificial intelligence it was predicted that AI will have an impact and will play a major role in the future, there will come a time that these AI will think and feel like human, I've seen these scenarios on futuristic movies.
I hate to say this but the developers of these AI seem dedicated to creating a perfect Artificial Intelligence that could possibly be used to predict sports betting accurately and helping developing other industries, beating humans to their jobs, am I the only one worried on this future scenario?