Carder.life - FAQ: How to Build an AI-Powered Telegram Bot Using Python & OpenAI API (2025)

? How to Build an AI-Powered Telegram Bot Using Python & OpenAI API
AI chatbots are becoming increasingly popular for automation, customer support, and cybersecurity. In this guide, I'll show you how to create a Telegram bot using Python, connect it to OpenAI's GPT-4 API, and deploy it on a server for 24/7 operation.
What this guide covers:
✔️ Setting up a Telegram bot with Python
✔️ Connecting it to OpenAI’s GPT-4 API
✔️ Deploying the bot on a VPS for continuous operation
✔️ Expanding features with web scraping, automation, and security
?️ Step 1: Creating a Telegram Bot
1. Open Telegram and search for @BotFather
2. Use the command:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 34px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">/newbot</pre>
3. Choose a bot name and a unique username (ex. 2crdforum_bot)
4. Copy the API token provided by BotFather (you'll need it later)
⚙️ Step 2: Installing Python and Dependencies
Install the necessary libraries:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 34px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">pip install python-telegram-bot openai requests</pre>
Create a new file bot.py and import the required libraries:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 146px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">import logging
import openai
from telegram import Update
from telegram.ext import Updater, CommandHandler, MessageHandler, Filters, CallbackContext
openai.api_key = "YOUR_OPENAI_API_KEY"
logging.basicConfig(level=logging.INFO)</pre>
? Step 3: Writing the AI-Powered Response Handler
Now, let's add the function that processes user messages and generates AI-powered responses:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 130px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">def handle_message(update: Update, context: CallbackContext):
user_text = update.message.text
response = openai.ChatCompletion.create(
model="gpt-4",
messages=[{"role": "user", "content": user_text}]
)
update.message.reply_text(response["choices"][0]["message"]["content"])</pre>
? Step 4: Connecting the Bot to Telegram
Now, add the main function to run the bot:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 162px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">def main():
updater = Updater("YOUR_TELEGRAM_BOT_TOKEN", use_context=True)
dp = updater.dispatcher
dp.add_handler(MessageHandler(Filters.text & ~Filters.command, handle_message))
updater.start_polling()
updater.idle()
if __name__ == "__main__":
main()</pre>
Run the bot with:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 34px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">python bot.py</pre>
? Step 5: Deploying the Bot on a VPS (24/7 Uptime)
To keep the bot running continuously, deploy it on a VPS like DigitalOcean, Linode, or Hetzner.
1. Install Python and required libraries:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 50px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">sudo apt update && sudo apt install python3-pip
pip install python-telegram-bot openai requests</pre>
2. Run the bot in the background:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 34px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">nohup python3 bot.py &</pre>
3. Use **screen** or **systemd** for better stability.
? Step 6: Expanding Features (Web Scraping & Automation)
Scrape News & Summarize with AI:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 146px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;"> import requests
from bs4 import BeautifulSoup
def get_latest_news():
url = "https://news.ycombinator.com/"
soup = BeautifulSoup(requests.get(url).content, "html.parser")
headlines = [a.text for a in soup.select(".titlelink")[:5]]
return "\n".join(headlines)</pre>
Now, integrate this function into the bot:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 82px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">def news(update: Update, context: CallbackContext):
update.message.reply_text(get_latest_news())
dp.add_handler(CommandHandler("news", news))</pre>
Now users can type **/news** to get AI-processed news updates.
?️ Security Considerations
? **Protect Your API Key:**
- Never hardcode API keys in your script.
- Store keys in a `.env` file and load them using `dotenv`.
⚡ **Rate Limit Requests:**
- Prevent abuse by limiting response length:
Code:
<pre class="alt2" dir="ltr" style="
margin: 0px;
padding: 6px;
border: 1px solid rgb(0, 0, 0);
width: 640px;
height: 66px;
text-align: left;
overflow: auto;
background: rgb(37, 37, 37) none repeat scroll 0% 0%;
border-radius: 5px;
font-size: 11px;
text-shadow: none;">if len(user_text) > 500:
update.message.reply_text("Message too long!")
return</pre>
? Conclusion
Now you have a fully functional AI-powered Telegram bot with GPT-4 support. You can expand it with automation, data scraping, and security features.
What AI features would you like to add? Share your ideas below! ?