Rexigon chess - new engine built with Kiro AI / Claude 3.5

[Graham Banks]

Hi, Tom.

At present, I'm inclined not to start testing a possible bunch of AI created engines. CCRL has Amira as the sole example in our lists, just out of interest.
That stance might perhaps change in future with a special list for such engines. Not sure.

Regards,
Graham.

[silentshark]

Hi, Tom.

At present, I'm inclined not to start testing a possible bunch of AI created engines. CCRL has Amira as the sole example in our lists, just out of interest.
That stance might perhaps change in future with a special list for such engines. Not sure.

Regards,
Graham.

Completely understand. Maybe, as you say a different list for these kinds of engines. Mind, where does it stop? We are on the cusp of being able to just ask AI - 'hey, please write me 5000 chess engines, all UCI compliant, using well known pruning techniques and eval. Make them all a bit different. Pop the source and executables into github. Give the engines cool and unique names.'. And then what do we do?

[silentshark]

A good game from Francesca vs. MChess in Paderborn (WM, many years ago)
Never forget it ... you made me and Kai very happy as operators.
Go Rexigon go ...

I'd never forget that! Still owe you and Kai a big thank you for operating Francesca all those years ago. And what a tournament. MChess was clearly a better engine, but in the late 90's we all discovered / were told about null move, and suddenly weak amateur engines could surprise the pros!

[jefk]

please write me 5000 chess engines, all UCI compliant, using well known pruning techniques and eval. Make them all a bit different. Pop the source and executables into github. Give the engines cool and unique names.'. And then what do we do?

run tournaments ?

[glav]

Hi, Tom.

At present, I'm inclined not to start testing a possible bunch of AI created engines. CCRL has Amira as the sole example in our lists, just out of interest.
That stance might perhaps change in future with a special list for such engines. Not sure.

Regards,
Graham.

Hi, Graham. First of all I would like to take the opportunity to thank you and the whole CCRL group for what you are doing for our community. If we thrive and keep trying new things, is certainly also due to the motivation we get from your lists.
Also, I must say that I respectfully disagree about not testing AI created engines. So far, there would be really very few program worth testing, since many of them do not even compile, crash or loose on time (mine!). However, even a dedicated list would be OK but, as I said, there is really a paucity of serious AI-generated programs to test.