The only solution is to choose/create a game that doesn't have any AI that plays it better than humans.
AI will always come in the lead for the mere fact that they are trained on/for a dataset that will be beyond human experience/capability. Dota 2, for example, has no "moves" to it. Everything is "Dynamic" yet OpenAI which was trained for a few weeks came forward and
beat the world's best[1]. Here is a quote from
OpenAI blog[2] regarding the timeframe the bot spent on learning.
In total, the current version of OpenAI Five has consumed 800 petaflop/s-days and experienced about 45,000 years of Dota self-play over 10 realtime months (up from about 10,000 years over 1.5 realtime months as of The International), for an average of 250 years of simulated experience per day. The Finals version of OpenAI Five has a 99.9% winrate versus the TI version.
Same happened with "Go", this game was thought to be something which AI's can never be good at, But then came "AlphaGo" and beat the best of us. Currently, AlphaGo has the lead. AIs given enough training time will always come in the lead as they are less prone to making errors and have tons of experience on their back. I believe there will be no game in which humans will come out top. AIs will always take the lead.
As for OP, I believe there is still a chance for his project. He could make this project for friends who want to play against each other. People who trust each other enough that they won't screw each other via cheating could use a service like OPs but then again the People who OP will target will be very small in numbers. Another way to use his project could be, OP could make this a faucet of sorts. Teaching people how to Play checkers and at the same time get rewarded... But if there isn't a way to sustain this project then it's better to kill/
sell[3] it than operate it at a loss.
Sources:
[1]
https://www.youtube.com/watch?v=yBEidvm_tZQ[2]
https://openai.com/blog/openai-five-defeats-dota-2-world-champions/[3]
https://flippa.com/sell