Airtable

Named native integrations for Slack, Google Drive, Salesforce, Jira and Zendesk with a full integrations directory and a public marketplace; agents call document analysis, web search and image generation tools.

Sourceairtable.com/integrationsread 2026-09-01

Omni is a conversational builder that configures the work rather than only answering: the help center documents it creating tables and fields including field agents, building interface pages and AI-generated interface elements, and creating or updating automations, all from a plain language request.

Airtable's own engineering blog describes a single agentic framework behind both Omni and Field Agents whose agents reason, plan and orchestrate actions to accomplish complex tasks with minimal input, and the agents page states agents orchestrate actions across the entire operation. Underneath sits a dedicated automations engine as a first-class platform module.

Sourceairtable.com/platform/ai-agents, support.airtable.com using-omni-ai-in-airtable and the Airtable engineering blogread 2026-09-01

Agents are grounded in the customer relational base and can read across connected data; document analysis agents scan contracts, invoices and reports into structured fields and web search agents pull real time external data into the same records.

Sourceairtable.com/platform/ai-agentsread 2026-09-01

Oversight is governed at the boundary: admins enable AI per workspace, customers control what data agents access, role based access control applies and all agent activity is auditable, but no approval gate or review before execute step on individual agent actions is documented.

Sourceairtable.com/platform/ai-agentsread 2026-09-01

Airtable documents both attestations and controls on its trust page. Certifications: SOC 2 Type 2 with the report available on request, ISO/IEC 27001:2022 and ISO/IEC 27701:2019 with certificates published, TX-RAMP Level 2 with certificate, and HIPAA, alongside GDPR, UK GDPR and CCPA/CPRA programs, a signable DPA and a maintained subprocessor list.

Controls: SAML single sign on and additional admin features on Enterprise, two-factor authentication, customizable collaborator permissions, view access restricted by password or email domain, SCIM provisioning and enterprise key management letting customers hold their own keys at rest. Additionally a HackerOne bug bounty program, published security and privacy whitepapers, and completed CAIQ, SIG Lite and HECVAT questionnaires available under NDA.

Sourceairtable.com/company/trust-and-securityread 2026-09-01

Three layers of record. Enterprise audit logs are available through the admin panel and through a dedicated Audit Logs API with published event types and history back to April 2021. Record-level revision history gives a visual activity feed of every change made to each record, which is where Field Agent output lands, so an agent's writes are inspectable at the row it wrote. Admin reports cover AI credit consumption, API and token owners and enterprise key management status. The vendor states directly on the agents page that all Field Agent activity is governed and auditable.

Sourceairtable.com developer audit log event types, airtable.com/company/trust-and-security and airtable.com/platform/ai-agentsread 2026-09-01

Agent outputs persist as structured fields in the relational base and become shared state other agents read, and HyperDB scales to 100M records, but no cross session agent memory or learned context layer distinct from the data store is documented.

Sourceairtable.com/platform/hyperdbread 2026-09-01

Airtable offers a real but narrow choice. The trust page offers customers the option to have their data stored in the European Union, with its own FAQ, and gives Enterprise customers enterprise key management so data at rest is held under customer-owned keys; models can additionally be served through Amazon Bedrock so providers never touch customer data. No self-hosted, on-premises, VPC or customer-cloud deployment exists, and there is no FedRAMP authorization, so the platform is cloud only with a residency option rather than a choice of deployment topology.

Sourceairtable.com/company/trust-and-security and airtable.com/company/data-residency-faqsread 2026-09-01

Packaged assets ship in three forms, all adoptable without building. AI Plays is a browsable library of premade AI workflows at airtable.com/ai-plays, each with its own page, including campaign concept ideation, brand and compliance review, localization of image and copy variants, and contract operations.

The agents page names four ready agent types, document analysis, web search, image generation and custom. And the help center documents a pre-set field agent catalog: choosing a suggested agent opens Omni with a request to build it, so a packaged agent is adopted in one step and then editable. Alongside these sit an extensive template gallery and a public marketplace.

Sourceairtable.com/ai-plays, airtable.com/platform/ai-agents and support.airtable.com using-airtable-ai-in-fieldsread 2026-09-01

Field Agents can be set to run automatically whenever data is added or updated so they work continuously without manual invocation; automations provide event triggers; web, desktop and mobile apps plus webhooks.

Sourceairtable.com/platform/ai-agentsread 2026-09-01

Enterprise customers decide which models are enabled and choose among OpenAI, Anthropic and Meta models; Anthropic Claude models are served via Amazon Bedrock so providers never handle customer data, and providers neither retain data nor train on it.

Sourceairtable.com/platform/ai-agentsread 2026-09-01

Web API, Enterprise API, Audit Logs API and Shares API with a developer portal, personal access tokens, webhooks, scripting and custom extensions, plus an official MCP server and skills exposing bases, records and workflows to external agents and a promoted Claude integration landing page.

Sourceairtable.com/developers/web/api/introductionread 2026-09-01

What the customer can inspect is the build, not the agent's behavior. Omni produces a working configuration the user reviews and then iterates on conversationally, and the help center documents its limits explicitly, including which interface layouts it cannot create, so the loop is build, look, adjust. Automations expose run history for debugging a failed step. No agent evaluation harness, no scored or repeatable test run, no regression suite and no output scoring surface is documented.

Sourcesupport.airtable.com using-omni-ai-in-airtable and airtable.com/platform/automationsread 2026-09-01

Web search agents retrieve real time web data into records, but no browser control, page navigation or computer use capability is documented anywhere on the product surface.

Sourceairtable.com/platform/ai-agentsread 2026-09-01