Recently, I've been talking to a lot of people, and everyone seems very pessimistic; indeed, liquidity is quite poor. I have a vague feeling that this is similar to before the DeFi summer, before the summer of 2020, when everyone believed that DeFi (back then, this term wasn't even a consensus) was the future, but no one knew how to proceed. Then in the summer of 2020, Compound, Aave, Curve emerged, with very simple data formulas that were elegant, effective, and profitable. Now, everyone knows that AI is the next biggest gold mine, but it seems like no one knows what to do or how to be effective; everyone is waiting for some extremely elegant, simple, and effective innovations to turn this innovation into attention and assets. At this point in time, it might be the best moment for AI entrepreneurs to start their ventures. The cost of AI has been continuously decreasing compared to last year, with a plethora of models and workflows available, and many effective outcomes can be achieved. The core is, front store and back factory; the product you package must meet certain needs and solve specific problems. The past Crypto + AI may need to transform into AI + Crypto: 1. AI to solve the imbalance of demand and supply 2. Crypto to solve the imbalance of payment and revenue At least one thing I am very certain about is that Crypto is suitable for creating AI interfaces; boring infrastructure is being debunked one by one, and it's time to look at interesting AI applications. The boundaries between 2 and 3 may become increasingly blurred, with AI applications that solve problems and generate real income being able to thrive on both sides. @GMAgents_AI We have been thinking about the opportunities for entry applications in AI and how to proceed. Friends who are working on AI applications are also welcome to exchange ideas and discuss together.