Airtable

Omni is Airtable's AI app builder. You describe what you need and it creates the tables, interfaces and automations for you, then you refine the result with normal drag-and-drop tools. You can attach briefs, meeting notes or docs so it tailors the app to your context. Asking it to build or iterate costs nothing extra.

An AI field runs a model over a row, so you can generate content, summarize text, categorize items, run sentiment analysis or translate copy. You start from a template or write your own prompt and reference fields the way you would in a formula. Outputs refresh automatically as records change.

Field agents go beyond a single field. They can analyze documents, search the web, generate and edit images, translate content and pull insights from transcripts, all within the record. You can set them to run on their own whenever data is added or updated. Each run costs credits.

You can stack AI fields so the output of one feeds the next. A first field scores customer sentiment; a second drafts a reply that matches the tone. Airtable's automations then send the result, which turns a manual review loop into a pipeline that runs itself.

Enterprise customers pick which model powers Airtable AI. Options include OpenAI's GPT models and Anthropic's Claude models through Amazon Bedrock, where the models run inside the AWS environment. That lets a company match the model to its own data and compliance needs.

Airtable connects to tools like Slack, Google Drive, Salesforce and Jira, and data moves both ways. Because the AI sits on top of that connected data, a summary or a tag can draw on records that came in from another system without a separate export step.