
Build Trading Strategies With Just One Sentence? Testing the New Binance AI Pro
The day before yesterday, @binancezh announced its new AI product ecosystem, Binance Intelligence, which includes Binance AI for regular users, Binance AI Pro for advanced users, and Agent OS for developers.
As a longtime user of Binance's AI products, I've followed everything from the initial Binance AI, to the later ChatGPT plugin integration, to the recent months of continuous feature additions focused on trading, data, and agents. Basically, every time a new version of Binance's AI product is released, I give it a try.
So this time around, seeing the new AI Pro, what I really want to know is: what exactly is different from before? And can it help me make money 24 hours a day?

🌟What is AI Pro?
According to the official breakdown, Binance Intelligence is primarily divided into three layers:
- Binance AI: For regular users, free to use. It can generate summaries for Crypto, US stocks, and macro markets, and will adjust content based on user usage, preferences, and holdings.
- AI Pro: Designed for more advanced users. You can describe trading ideas in natural language, and it will convert them into an automated workflow.
- Agent OS: Built for developers, enabling external AI Agents to connect with Binance's data and trading capabilities.
During the press event, Binance CEO Richard Teng @_RichardTeng and Product Vice President Jeff Li @taida88 introduced AI Pro by explaining that users simply need to describe their trading idea in one sentence, and AI Pro will organize it into a workflow. It allows for revision and simulation first, and enters live trading only after confirmation.
"Workflow" was also a highly frequent term throughout the presentation. You can directly tell the AI under what conditions to open positions, which indicators to watch, and how to manage position sizes, and it will break those conditions down into a strategy. After generation, you can continue to modify the strategy, test it first in a simulated environment, and finally decide whether to run it live. The official process is summarized as Prompt, Test, Trade.
Beyond trading, the presentation also demonstrated scenarios such as automatically generating daily market reports, continuously monitoring BTC, and sending alerts at key price levels.
After watching the event, I have a general sense of AI Pro, so next I'll directly look at how it performs when applied to real-world needs.
🌟Hands-on Testing: What Can AI Pro Do
What I really wanted to verify was how these features perform in actual trading scenarios, and whether they are truly as impressive as presented at the event.
After all, for trading AI, being able to chat is no longer novel. More importantly, can it truly understand my requirements, pull real-time data and account information, and further help me generate strategies and execute tasks?
So this time, I picked two scenarios I actually use regularly: wealth management and trading strategies. The former tests whether it can reduce the time spent filtering products, while the latter directly tests the core capability of the new AI Pro.
1️⃣Where should I put 50 BTC? Let AI Pro choose for me first
Previous versions of Binance AI and the ChatGPT plugin could already check market prices and view data, so I didn't want to repeatedly test those basic capabilities this time. Instead, I started with a more practical scenario.
Fortunately, the new AI Pro is already integrated with Binance Earn, so I asked directly:
I have 50 BTC that I want to place in Binance Earn to earn yields, but I don't want to bear the risk of principal loss. How would you allocate it? Please prioritize recommending schemes with higher current yields and lower risks.
AI Pro directly calls the current product data from Binance Earn, while also reading my actual account balance.
Since my account actually holds very little BTC, it even proactively reminded me: "You currently do not have 50 BTC in your spot account. If you really intend to execute, you will need to transfer the BTC into your account first."
In product selection, it also understood my core requirements, proactively excluding Dual Investment and other products that could change the BTC amount, and prioritizing recommendations for BTC flexible and fixed-term options. It found that the BTC flexible APR was approximately 0.26% at the time, along with a special fixed-term product offering up to a 20% APR over 5 days, which basically matched what I later saw on the Earn page.

2️⃣Trading Strategy Test: Can It Understand My Trading Logic?
Next comes the feature I most wanted to try—creating a strategy.
This time, I deliberately didn't give it a simple "buy on breakout." Instead, I threw in several types of data I normally use for futures trading to see if it could still sort out the logic once the conditions became complex.
My prompt to it was:
Help me create a BTCUSDT perpetual long strategy: 1h and 4h both above MA20; 15m reclaims MA20 and volume exceeds 1.3 times the average of the past 20 candles; OI increased by over 5% in the last 1h, active buy/sell ratio >1, funding rate <0.01% triggers entry. Position size 10%, 2x leverage, stop-loss 1.5%, take-profit 3%. Generate the strategy first, do not trade live.
Its performance this round was actually better than I expected.
It broke down the entire logic quite comprehensively: 1h and 4h determine the larger timeframe direction, 15m looks for entry points, volume confirms the breakout, OI and active buy/sell ratio monitor capital synchronization, and Funding filters out overheated market conditions.
Details at the execution layer were also handled well. For example, regarding whether "position size 10%" refers to notional position or margin percentage, it proactively explained its understanding. Parameters like isolated mode, one-way position, and checking frequency, which weren't explicitly stated, were also brought up in advance for user confirmation.
After confirming the strategy, AI Pro can also run it in a simulated environment first. I directly put this strategy into simulated trading, letting it continuously check conditions against real-time market data. This way, on one hand, I can see if the strategy triggers correctly, and on the other hand, I can identify any logical or parameter issues beforehand.
Currently, this strategy is running in the simulated environment. Later, if the performance remains stable and the trigger logic meets expectations, I will consider switching it to live trading.

🌟Final Thoughts
After these two rounds of testing, my impression of the new AI Pro is actually better than expected.
During previous research on AI Trading, I increasingly felt that AI's advantage lies in monitoring the market 24/7, checking data, waiting for conditions, and executing strategies. Humans get tired and lose focus, but AI does not. This is precisely where it excels in trading, which is also why more and more trading tools are recently shifting toward the AI Agent direction.
Meanwhile, Binance has a clear advantage in doing this: Binance boasts the deepest liquidity, a massive user base, and a large amount of first-hand trading data. Many third-party AI tools still need to independently integrate data feeds, accounts, and trading APIs, whereas Binance can directly embed AI into its existing trading infrastructure.
So, what strikes me most about AI Pro this time is that AI is beginning to truly enter the trading era.















