ChatGPT Integration • page 3/3 • General Chess Discussion • lichess.org

@Brian-E said in #20:

The big difference between ChatGPT and search engines like Google is that with the search engine you see the source. You also see multiple different options. ChatGPT, as I understand it, uses similar methods to a search engine but fails to provide the source of what is it saying and gives just one potted response.

Well sure, that just comes down to its mechanics but at this point we are off topic. We can talk more about mechanics but it’s seems that we simply just have another vote for not using Chat for its brains in chess but rather as a voice and to use Stockfish as the brain

@Brian-E said in #20: > The big difference between ChatGPT and search engines like Google is that with the search engine you see the source. You also see multiple different options. ChatGPT, as I understand it, uses similar methods to a search engine but fails to provide the source of what is it saying and gives just one potted response. Well sure, that just comes down to its mechanics but at this point we are off topic. We can talk more about mechanics but it’s seems that we simply just have another vote for not using Chat for its brains in chess but rather as a voice and to use Stockfish as the brain

corvusmellori

#22

If someone actually wanted to try this project, probably the best way would be to use the old Hand-Crafted Evaluation, non neural-net version of stockfish. This gives points for things like king safety, piece mobility, and so on, and someone who was motivated to could build a tool that exposed all the HCE components in a way that was explainable.

pointlesswindows

#23

"So I've been playing chess for a while and still have my areas of improvement"

yup

"So I've been playing chess for a while and still have my areas of improvement" yup

Ifancy_potato

#24

I had chatGPT write this for me to see what he thinks about the topic.
I'd like to share my thoughts on the topic of integrating AI into Stockfish chess bots on Lichess, and why I believe it's not necessarily a good idea. While AI can certainly offer many advantages, there are some potential drawbacks and concerns to consider:

"Unbalancing the Playing Field: Stockfish, even in its current form, is already an incredibly powerful and respected chess engine. Integrating AI could make it nearly unbeatable for the average player. This might discourage many from playing against it, ultimately reducing the appeal of human vs. AI games.

Loss of Human Essence: Chess is a game that has been enjoyed by humans for centuries. Adding too much AI sophistication to Stockfish could lead to a loss of the human essence in chess. Part of the beauty of the game lies in its imperfections and unpredictability, which AI could diminish.

Accessibility and Learning: Chess engines like Stockfish have been valuable tools for players to learn and improve their skills. If AI integration makes Stockfish too complex or unpredictable, it could become less suitable for educational purposes, as players might not understand the reasoning behind its moves.

Ethical Concerns: AI integration can raise ethical questions. How much AI is too much? When does it cease to be a chess bot and become a superhuman entity? These questions could create discomfort and divisions in the chess community.

AI Arms Race: There's also the risk of an AI arms race, with different chess platforms striving to make the strongest AI-powered bots. This could lead to an unhealthy competition where the emphasis shifts from chess strategy and skill to AI development and implementation.

Reduced Human vs. Human Interaction: Chess is not just about winning or losing but also about human interaction, competition, and camaraderie. If AI-powered Stockfish becomes the norm, it may reduce opportunities for players to engage in human vs. human matches, which are at the heart of chess culture.

Compatibility and Accessibility: Integrating AI might make Stockfish less accessible for players with older or less powerful devices. This could potentially exclude a portion of the Lichess community from enjoying the game.

In summary, while AI integration has its merits and could lead to exciting developments in chess, it's essential to consider the potential drawbacks and strike a balance between technological advancement and preserving the core values of the game. Let's keep Stockfish strong but ensure that it remains an enjoyable and educational experience for all chess enthusiasts."

I had chatGPT write this for me to see what he thinks about the topic. I'd like to share my thoughts on the topic of integrating AI into Stockfish chess bots on Lichess, and why I believe it's not necessarily a good idea. While AI can certainly offer many advantages, there are some potential drawbacks and concerns to consider: "Unbalancing the Playing Field: Stockfish, even in its current form, is already an incredibly powerful and respected chess engine. Integrating AI could make it nearly unbeatable for the average player. This might discourage many from playing against it, ultimately reducing the appeal of human vs. AI games. Loss of Human Essence: Chess is a game that has been enjoyed by humans for centuries. Adding too much AI sophistication to Stockfish could lead to a loss of the human essence in chess. Part of the beauty of the game lies in its imperfections and unpredictability, which AI could diminish. Accessibility and Learning: Chess engines like Stockfish have been valuable tools for players to learn and improve their skills. If AI integration makes Stockfish too complex or unpredictable, it could become less suitable for educational purposes, as players might not understand the reasoning behind its moves. Ethical Concerns: AI integration can raise ethical questions. How much AI is too much? When does it cease to be a chess bot and become a superhuman entity? These questions could create discomfort and divisions in the chess community. AI Arms Race: There's also the risk of an AI arms race, with different chess platforms striving to make the strongest AI-powered bots. This could lead to an unhealthy competition where the emphasis shifts from chess strategy and skill to AI development and implementation. Reduced Human vs. Human Interaction: Chess is not just about winning or losing but also about human interaction, competition, and camaraderie. If AI-powered Stockfish becomes the norm, it may reduce opportunities for players to engage in human vs. human matches, which are at the heart of chess culture. Compatibility and Accessibility: Integrating AI might make Stockfish less accessible for players with older or less powerful devices. This could potentially exclude a portion of the Lichess community from enjoying the game. In summary, while AI integration has its merits and could lead to exciting developments in chess, it's essential to consider the potential drawbacks and strike a balance between technological advancement and preserving the core values of the game. Let's keep Stockfish strong but ensure that it remains an enjoyable and educational experience for all chess enthusiasts."

Panagrellus

#25

There are already a couple of projects combining language AIs and chess engines to build an "AI coach" that speaks human language. Not sure if they are any good though - e.g. noctie.ai comes out as the first google hit for me.