ChatGPT is all the rage… but can it succeed at poker? More importantly, can it do the impossible, and explain how to play Texas Hold ‘Em? Like, make the game so easy to understand even a human can do it? Let’s find out.
I decided to go in strong, and challenge the artificial intelligence machine to outline one of the trickiest concepts in poker: Position.
“Explain position in poker in very simple terms,” I told the program. Here’s what it had to say:
In poker, position refers to where a player is seated at the table in relation to the dealer. The player who is closest to the left of the dealer is considered to be in "early position," and the player who is closest to the right of the dealer is considered to be in "late position." The later a player's position, the more information they have about the other players' actions, which can give them an advantage in making decisions about whether to bet, call, or fold. This is because the players in later position get to act after the players in earlier positions, and can therefore see what decisions they make before they have to act themselves.
Even at first glance, that explainer seemed quite long… like a wall of text. Who wants to read all that? Already the answer was too dense for my taste.
So I tried again.
“Explain it more simply,” I commanded. It replied:
In poker, position refers to where you are sitting at the table. If you're sitting to the left of the dealer, you're in an "early" position. If you're sitting to the right of the dealer, you're in a "late" position. Being in a later position is generally better because you get to see what the other players do before you have to make your own decision. This can help you make better decisions about whether to bet, call, or fold.
Still a bit long, but better. Why didn’t it just say this in the first place? Also, personally speaking, I could’ve done without the last line. It seems repetitive and unnecessary. Edit yourself, Mr. GPT!
Next I decided to turn away from strategy and see if I could get closer to an opinion. “Please list the best poker players in the world and what makes them unique,” I prompted, to which it diplomatically replied:
As an AI language model, I don't have personal preferences or opinions, but here is a list of some of the most accomplished and well-respected poker players in the world, along with a brief description of what makes them unique:
- Phil Ivey: Known for his incredible poker instincts, ability to read opponents, and aggressive playing style. He has won 10 World Series of Poker bracelets and is considered one of the best all-around players in the game.
- Daniel Negreanu: Known for his friendly demeanor and ability to make tough decisions under pressure. He is a master of reading opponents and has won 6 World Series of Poker bracelets.
- Doyle Brunson - A legendary player who helped popularize Texas Hold'em, Brunson has won 10 World Series of Poker bracelets and is known for his aggressive playing style and ability to make big bluffs.
- Phil Hellmuth - Known for his brash personality and flashy entrances to poker tournaments, Hellmuth has won 15 World Series of Poker bracelets and is a master of the "white magic" of poker, which involves making strong plays with weak hands
- Erik Seidel - A veteran player who has been successful in both live and online tournaments. He is known for his analytical mind, ability to adapt to changing circumstances, and his cool demeanor at the table.
- Fedor Holz - One of the most successful young players in the game today, Holz has won over $32 million in live tournament earnings and is known for his mathematical approach to the game and his ability to stay focused and composed under pressure.
- Vanessa Selbst - A retired professional player who was known for her aggressive style and ability to read opponents. She has won three World Series of Poker bracelets and is considered one of the best female players in the game.
- Chris Moneymaker - A recreational player who won the 2003 World Series of Poker Main Event and helped popularize the game to a wider audience. He is known for his calm demeanor and ability to make strong plays under pressure.
- Antonio Esfandiari - A former magician who is known for his outgoing personality and ability to make big bluffs. He has won three World Series of Poker bracelets and is considered one of the most entertaining players in the game.
- Johnny Chan - A two-time World Series of Poker Main Event winner, Chan is known for his aggressive playing style and ability to read opponents. He was a dominant force in the game during the 1980s and 1990s and is still respected by players today.
Overall this was more to my liking. Phil Ivey is characterized by his “incredible poker instincts,” Daniel Negraenu by his “friendly demeanor,” and Phil Hellmuth by his “brash personality and flashy entrances.” But many of the descriptions were repetitive… and extremely outdated. I know ChatGPT’s search engine isn’t updated to include information up to 2023, but can’t we get some reference to Kristen Bicknell or her husband Alex Foxen? Some of these names haven’t played a televised poker tournament in ages.
It was time to really push the software.
“Why are most people bad at poker?” I asked with a snicker.
Most people are not necessarily "bad" at poker, but rather they lack the experience, knowledge, and discipline required to consistently make good decisions at the poker table. Poker is a complex game that requires a combination of skill, strategy, and luck. In order to become a successful player, you need to have a deep understanding of the rules, different betting strategies, hand rankings, and be able to read your opponents.
Additionally, many people approach poker with the wrong mindset. They may be too focused on trying to win every hand, or they may become too emotionally invested in the game and start making poor decisions. Successful poker players need to be able to maintain a calm and analytical mindset, even when faced with setbacks or losses.
Finally, many people simply do not invest the time and effort required to become proficient at the game. Becoming a successful poker player takes a lot of practice, study, and hard work. If you're willing to put in the time and effort, however, it's possible to improve your skills and become a better player over time.
Ah, now we are getting somewhere! That first sentence was actually genuinely insightful (and a bit of a hot take). “Most people are not necessarily 'bad' at poker,” countered my premise, and the follow-up, they “rather they lack the experience, knowledge, and discipline required to consistently make good decisions at the poker table” made me think twice. [Key word: consistently].
Apparently the program doesn’t seem to believe in talent or natural affinities as much as it believes in developing a skill set.
I wanted to learn more about ChatGPT’s views on poker.
“Should poker be taught in schools?” I asked, to which it boringly replied:
The decision to include poker as part of a school curriculum is a complex issue that depends on many factors, including the age of the students, the culture and values of the community, and the specific goals of the educational program.
On one hand, there are several benefits to teaching poker in schools. Poker requires a combination of math skills, strategic thinking, and decision-making abilities, all of which are important for success in many other areas of life. Additionally, poker can help teach important life skills such as risk management, emotional regulation, and discipline.
However, there are also concerns about promoting gambling or exposing students to potentially addictive behaviors at a young age. It's important to carefully consider the potential risks and benefits of introducing poker into a school curriculum, and to ensure that any educational program is age-appropriate, responsible, and aligned with the values and goals of the school and community.
Ultimately, the decision to teach poker in schools should be based on a careful consideration of the potential benefits and risks, and should be made in consultation with parents, educators, and community stakeholders.
This response sounded so canned and overly neutral. So much for another point-of-view!
I needed to lighten the mood a little bit.
“Tell me a poker joke!” I queried, to which it replied:
Sure, here's a classic poker joke:
"Why did the poker player go broke? He lost all of his chips!"
It's a bit of a cheesy pun, but it's a classic that's sure to get a chuckle from fellow poker enthusiasts.
The pun doesn’t make much sense, but I genuinely chuckled in a “it’s-so-bad-it’s-good” kind of way. Sometimes reality has its own dark humor.
“What do you do if someone at a poker table asks if you are bluffing?” I then asked the program, wanting to test an AI’s people skills.
If someone at a poker table asks you if you are bluffing, it is generally not advisable to provide a direct answer. Revealing whether or not you are bluffing can give your opponents valuable information about your hand and your playing style, potentially giving them an advantage in the game.
Instead of answering the question directly, you can try to deflect or avoid the question by making a non-committal statement, changing the subject, or simply remaining silent. You can also use the opportunity to observe the other players and gather information about their reactions and behavior, which can help you make better decisions in future hands.
It's important to remember that bluffing is a key strategy in poker, and it is natural for players to try to figure out if their opponents are bluffing. However, revealing too much information about your own hand or strategy can be detrimental to your success in the game. So, it's usually best to avoid answering directly when asked if you are bluffing and instead focus on making strategic decisions based on the information available to you.
Not bad! ChatGPT seemed to realize that when it comes to cards, honesty isn’t always the best policy. Could ChatGPT be a bit of a shark?
It was time to find out and ask the ultimate question: “Are computers better than people at poker?”
Here’s what it had to say:
In many ways, computers are better than people at poker. They are able to analyze vast amounts of data and calculate probabilities and optimal strategies with incredible speed and accuracy, something that is beyond the capabilities of even the most skilled human players. Additionally, computers are able to play with perfect memory, meaning they can track every card that has been played and use this information to make more informed decisions.
Computer programs that have been specifically designed to play poker, known as "poker bots," have been shown to be able to beat even the best human players in certain types of games and situations. For example, in 2017, an artificial intelligence program called Libratus was able to defeat a group of top human players in a 20-day poker tournament.
However, there are still some areas where human players have an advantage over computers. Poker is a game that involves a significant amount of psychology and human interaction, including bluffing, reading opponents, and detecting patterns in behavior. Human players are often better able to pick up on these subtle cues and adjust their strategies accordingly.
Additionally, computers are only as good as the algorithms and data they are programmed with, and they may struggle to adapt to new or unexpected situations. This is why many experts believe that the best poker players of the future will be those who are able to combine the strengths of human intuition and adaptability with the analytical power of computer algorithms.
Wow… tell me how you really feel! I finally got ChatGPT to show its true colors. “In many ways, computers are better than people at poker,” it starts. Yes, yes, towards the end it throws a bone towards “some” areas where humans “still” have an advantage but that part is buried near the end and laced with doubt (human players are “often” better able to pick up on these subtle cues, it says, bolding and quotations mine). That first sentence saying in many ways computers are better than people at poker seems less like a cheeky intro and more like a confident thesis statement.
Obviously, I was extremely offended! How dare it say such a thing! But ChatGPT forgets a staple of some of the best hustlers around: Keep the whales happy. Make sure the loose, fun, “action” players don’t want to leave the table. Well I had personally had enough! It was time to close the window. Bye, ChatGPT. I’m cashing out.
