How to Intermediate · 4 min read

How to use AI to process emails automatically

Q: How to use AI to process emails automatically

Use an AI language model like gpt-4o to automatically read and classify emails by extracting key information and generating responses. Integrate with email APIs (e.g., Gmail API) to fetch emails, then send the email content to the AI model for processing and automation.

Quick answer

Use an AI language model like gpt-4o to automatically read and classify emails by extracting key information and generating responses. Integrate with email APIs (e.g., Gmail API) to fetch emails, then send the email content to the AI model for processing and automation.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0
Access to an email API (e.g., Gmail API) with OAuth2 credentials

Setup

Install the OpenAI Python SDK and set your API key as an environment variable. Also, set up access to your email provider's API (like Gmail API) to fetch emails programmatically.

bash

pip install openai google-auth google-auth-oauthlib google-auth-httplib2 google-api-python-client

Step by step

This example fetches unread emails from Gmail, sends the email content to gpt-4o for summarization and classification, then prints the AI-generated summary and suggested action.

python

import os
from openai import OpenAI
from google.oauth2.credentials import Credentials
from googleapiclient.discovery import build

# Initialize OpenAI client
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Load Gmail API credentials (assumes token.json exists from OAuth flow)
creds = Credentials.from_authorized_user_file('token.json', ['https://www.googleapis.com/auth/gmail.readonly'])
service = build('gmail', 'v1', credentials=creds)

# Fetch unread emails
results = service.users().messages().list(userId='me', labelIds=['UNREAD'], maxResults=5).execute()
messages = results.get('messages', [])

for msg in messages:
    msg_data = service.users().messages().get(userId='me', id=msg['id'], format='full').execute()
    parts = msg_data['payload'].get('parts', [])
    body = ''
    for part in parts:
        if part['mimeType'] == 'text/plain':
            import base64
            body = base64.urlsafe_b64decode(part['body']['data']).decode('utf-8')
            break

    prompt = f"You are an email assistant. Summarize this email and suggest a category (e.g., urgent, spam, personal, work):\n\n{body}"

    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": prompt}]
    )

    print("Email summary and category:")
    print(response.choices[0].message.content)
    print("---")

output

Email summary and category:
This email is a meeting request for next Tuesday regarding the Q2 project update. Category: Work
---

Common variations

Use gpt-4o-mini for faster, cheaper processing with slightly less accuracy.
Implement asynchronous calls with asyncio for handling large volumes of emails.
Stream responses for real-time processing using streaming endpoints if supported.
Integrate with other email providers by adapting the API calls (e.g., Microsoft Graph API for Outlook).

Troubleshooting

If emails are not fetched, verify OAuth token permissions and refresh tokens.
If the AI returns irrelevant summaries, refine the prompt with clearer instructions.
For rate limits, implement exponential backoff or batch processing.
Ensure environment variables are correctly set to avoid authentication errors.

✅

Key Takeaways

Use gpt-4o to extract summaries and classify emails automatically.
Combine email provider APIs with AI for end-to-end email automation workflows.
Refine prompts to improve AI understanding and output relevance.
Handle API rate limits and authentication carefully for reliable automation.

Verified 2026-04 · gpt-4o, gpt-4o-mini

Verify ↗