Topic: Can Chat GTP destroy poker on-line gambling??? - page 6. (Read 4814 times)

i don't think AI will ruin the world of poker gambling. their capability is different with real experienced poker players. i tried using chatgpt in a complex engineering prob, and it won't give you the answer that you want. however, it may give you important points how to solve a problem, but i can say, it has still limitations on what it can do. so the strategies acquired by a poker expert won't be matched by AI in my opinion. but some of its moves may be overwhelming to some especially those who are amateurs in this game.

I thought as much too, but Fast forward to today, we have seen the what extent AI has been designed. You barely even know what’s real anymore. There’s pictures, videos, simulated conversations, etc with AI. And for this reason, I think that it’s actually possible that we’ve met the systems that beat human nature, hence to my opinion that maybe, Chat GPT can destroy poker online gambling.

What is even the level of the independent of AI?

Can AI perform it task effectively without human involvement,  the answer is know so that point to the high dependence of AI on human contribution for it effective usage,  so for,  poke as an involving games that require skills to work around it made it a professional game with so many incredibly highly skilful players and for that AI dominance may still depend so much on the poker player who chooses to use them to achieve his goals in the game.

Quite agree with you mate, there is a popular saying In my place that "when a bird learns to fly without perching, the hunter will learn to shoot it without missing".
There is no problem in this world without solution except death.
Poker as a game as been for ages, and will still remains for ages to come, happily, the game have alot of incredible players who are so talented to know when something is wrong or unrealistic, this is why I personally agree with you that even if Ai usage becomes dominant in poker, it will always under the Control of man, the highly experienced players.

No. As I said, it is not.

Let's have citations directly from the paper.



You can skim the actual paper, you don't have to necessarily to read it carefully. The 5,000 and 10,000 plays are only to measure the effectiveness of the AI. It does not imply those are the number the AI learn the opponents play, they already have the gameplay blueprint/strategy right before it is rigorously tested.


Only by then, on the actual live gameplay, do they adapt and adjust accordingly. So it can't be said and is not to study the opponent's play by hours.

  I'm going to read it carefully this time. Thanks. But from the first glance, I see: "Playing more than 5,000 hands each time", "... for 10,000 hands"... Doesn't that mean "studying the opponent plays for hours"? I mean, to me, it does.
  
  But, as I said, although I read those articles 3-4 years ago, I think I should read them more carefully this time. Thank you again for the links.

You and others are right to worry about AI's rapid development. Knowing that AI brains calculate faster and smarter than human brains makes us concern.

However, considering AI awareness as human consciousness is conceptual. Many believe AI should be limited to prevent it from expanding too much. They are fast, accurate, and good at computer-intensive jobs. This makes them ideal for complex operations.

We must tread carefully on this path. Your emphasis on human control of machines is crucial. We must ensure their use benefits people and doesn't hurt them. Both good and negative possibilities exist, so we must be smart

Well so far things have not gone beyond normal, human intelligence is much more logical, so far we are better, it is obvious that things with AI can calculate faster, mathematics, the physical part, AI thinks more quickly and they obtain the results in seconds, in addition it is that we must make a calculation because we can do it quickly without however it will take longer, there is something that the AI cannot win, otherwise things can be very decisive because obviously Human Intelligence has some thoughts of another style, an AI that is talking a lot, it is the one that even spoke with CR7, I don't remember its name, but up to now it is already saying that it has consciousness as a human, and it believes that it is already a human, I don't know but I think that human beings must set limits on these things, because they can get out of control, all those movies that are AI where they take over the world, where they recreate possible keys to nuclear weapons, all those things can happen, it is not so far away anymore. From that, then the human being has to have control over these machines, because we don't know how much consciousness they can take on, go crazy, activate many things, dangerous bombs, attack countries, unleash wars, so this is not so far from the Actually, maybe not right now, because the AI is in a Beta phase, but later yes.


As far as I'm concerned, these things have been advancing a lot, there are programs that have robots learning certain types of things, they have them training, so this can be good or bad, and it's good because it would be great if it could use AI to do Beautiful things, Like very digital operations such as hand operations, brain operations, a lot of precision is required, and a well-trained AI with a mechanical hand can do it, so this is something that can be used, in South Korea they have very advanced medicine, maybe you see a good mechanism Action with the IAS can continue to raise the level of this , first of all it should focus on this type of work.

If that were the case, I am sure that investigation will be done. Look at Chess tournaments, Chess's got the most advanced AI in there already and even when people are using them to cheat, they can always detect them because experienced players know when something's too good to be true which is the likely scenario that would happen if AI will be used by someone to cheat in a poker tournament especially if you use it on amateurs, you will definitely become a suspect because you're playing too good.

AI may work at some point but it’s not completely reliable when it comes to gambling or let’s say online poker. The works of AI is still a lot different from a human being intelligence. Probably AI may somehow work but maybe it’s more on chances rather than using our human mind which creates more probability to win even in a very unpredictable gambling games.

I wouldn't go so far.

If I'm not wrong, skills count in Poker as much as 80%, and chance the remaining 20%. If a player had to compete against the machine, with no help, I don't think anyone would be able to calculate the probabilities in such a way to defeat the AI in the long term.

For games like chess where skill counts 100%, the probability for any human being to win taking into account the state of the art is zero, regardless his skill. With poker it is not zero because of the randomness you mentioned, but I still think that chances to win would be reduced dramatically.

Unless protocols like the ones mentioned by 3kpk3 can be successfully implemented, I think that AI can pose a threat for poker online in the way we know it.

Aside from that poker game has many variables for the AI to comprehend thoroughly.  We have the players skills, the randomness of the cards on deck and on hand, and many more.  I used AI to decide about my pokeplay, all it gives are just hints and suggestions which I mostly fold when asked giving the current cards in my hand.  So how can this kind of AI dominates the  Poker online gambling when it doesn't give me a good strategy to beat my opponent.

the only way AI can beat the game is when it is connected to the game server and cheats its opponent by knowing the variables including the cards on deck and on the player's hands.

The casino always has an advantage and the team also always monitors every customer activity while playing in one of the casino game options and from here it will be easy to find out if there are customers who use tools such as AI or ChatGPT to win the game and the team can immediately take action by How to stop an account or freeze it.
Only those who are obsessed with winning and want to make gambling a source of profit will carry out such actions because they no longer have confidence in their own abilities so they decide to use Al or ChatGPT.

Well AI is not a completely perfect technology because there are several shortcomings of AI that its users may not be aware of.

Al was created by humans with the aim of making things easier or helpful when there are things that cannot be known or solved so Al is said to be artificial intelligence but unfortunately there are many people who abuse Al and overestimate the advantages of Al in the gambling.
In my opinion those who believe too much that Al can help with everything in gambling are people who cannot think logically and don't want to do some learning so they can hone their gambling skills.

In my perspective as an IT graduate, I often use chatGpt for my work for reference only meaning I don't rely on it. Chat Gpt destroying poker of other online gambling? I think nope that's not gonna happen "for now" in the future maybe. You see AI technology is getting more advanced in our times but not that advanced that it can do anything. Of course owners or developers of certain online gambling games anticipated the advancement of the technology so they will do a lot or tight security in their system to ensure that AI will not be a problem. So rest assured ChatGPT doesn't have the capabilities to be used for online gambling.

Same had been said for chess when IBM defeated Garry Kasparov. I think it is just another step in evolution at least better to think that way.

Obviously, it is specifically for Pluribus, but surely the others also behave a similar way, not requiring analyzing the specific opponent's gameplay first. Let's refer to it from the Pluribus paper itself.

The main strategy that Pluribus employ is to use self-play (plays the game with copies of themselves). DeepStack and Libratus also use it. The result of the training strategy is used to play the real game, in the meantime, also adjusting to the state of the game to seek the best move.[1]

The way how works is resource intensive, it requires "~completing its blueprint strategy in eight days using just 12,400 core hours and 28 cores during live play." and since it is a successful algorithm, the researcher decided not to publicly publish the code.[2]

[1] https://www.science.org/doi/10.1126/science.aay2400#sec-2
[2] https://arstechnica.com/science/2019/07/facebook-ai-pluribus-defeats-top-poker-professionals-in-6-player-texas-holdem/

Well, I doubt that online casinos will be closed because they are experiencing losses due to many gamblers using AI. Casinos will develop AI that can withstand AI attacks from gamblers so that casinos can maintain their business and still make a profit. For casinos that do not have AI or do not develop AI technology enough in their business they will not be able to survive the ever-evolving AI and will close due to bankruptcy.

For now, it may be difficult for us to imagine what kind of AI gamblers will use. Maybe the AI technology is still under further development, so we need to wait for updates from the developers. And when there is someone who really understands the algorithms in casinos, they will try to continue to develop their AI so they can take advantage of casinos that don't make enough updates to their systems.

Yes, I'm aware of the existence of the machines you've mentioned, Pluribus, DeepStack, Libratus. But from what I know, they can't start beating anyone right away. They need to study the opponent plays for hours before they can start beating the professionals. And even theoretically I can't imagine how it can be otherwise. Do you have any link link proving your point?

As we know today's world has improved a lot and everything has changed, now many things can be done by AI. By AI now various tasks are done easily and quickly doing one job due to which people get a lot of benefit from using AI. But I think using this AI ChatGPT, maybe it is not possible to predict which casino games and how to play to become champion. And if it was possible to predict the casino by this AI, then surely the casinos would have been threatened and their gambling platforms would have been shut down. If on-line casinos could make predictions like this, then no one would bet in online casinos, but everything would be done in reality, because if it was successful, on-line casinos would have closed down. In that case I think it will never be possible, and this world master AI will never make all predictions of online casinos, as much as I think it is impossible.

Casinos have already anticipated this AI, in fact they are certainly much more sophisticated in dealing with users who try to use AI, so it will not be possible to give opportunities and wins to users who try to destroy it, after all AI technology cannot fully be used to earn money by beating casino, because of course that would also seem difficult.

Not all AI works like what we think, even if it is beyond the limits of our minds, AI does not necessarily provide that information, because human artificial intelligence will never be able to exceed what is beyond the limits of the human mind itself and in any case it is still needed. human touch is also used for gambling and does not fully use AI in playing online poker gambling.