Article

July 2025 Launch: Control and Context for AI Readiness

•

July 16, 2025

•

Summarize this article with:

At Airbyte, we believe flexibility and control are complementary. As the open standard in data movement, we're committed to giving people and organizations the freedom to focus on what matters: building, creating, and innovating with data as their foundation. Today, we're excited to announce our July 2025 launch, delivering the sovereignty, extensibility, speed, and reliability that modern data teams need to shape their future.

This launch addresses three critical challenges facing data organizations: governing data platforms with global data sovereignty requirements, preparing comprehensive data foundations for AI, and reducing the cost and complexity of data movement.

Sovereignty for Global Data Platforms with Multiple Data Planes

For organizations operating across multiple regions, data sovereignty has always been a complex puzzle. Physical geographic limitations and regulatory requirements have traditionally forced teams to implement intricate architectures with multiple control planes and data planes: a recipe for operational headaches and compliance risks.

We're changing that narrative with a solution that gives organizations complete control over how and where their data moves.

‍

Starting today, Airbyte Enterprise can be configured to orchestrate multiple data planes, each moving data across various regions all while allowing for easy governance from a single pane of glass. Our refreshed UI now allows you to assign data planes to workspaces directly from Airbyte's Workspace Settings. This simplifies data sovereignty across multiple regions, which in turn strengthens data democratization as your business scales globally.

To get started, simply deploy a data plane in a new Kubernetes environment, then configure it to register with the control plane. You will then be able to configure workloads to run in your newly created regions.

This is true sovereignty: data movement governed on your own terms, without compromise. Read the full breakdown here.

Move Structured and Unstructured Data Together to Achieve AI Readiness

AI is only as good as the data it ingests. Yet most data platforms force an artificial separation between structured records and unstructured files, creating gaps in context that AI models struggle to bridge. Our open, modular architecture makes it effortless for teams to move multimodal data no matter the situation, including the complex requirements of AI workloads.

Airbyte now enables you to move structured records and multimodal data (unstructured files, PDFs, videos, and more) together in the same connection. This extensibility means:

Tight relationships between content and metadata context
Preserved user permissions and identity metadata tied to replicated documents
Flexible file transfer modes with options to replicate records or copy raw files

‍

With these improvements, you can more easily engineer data platforms which provide your AI models with access to customer transaction records alongside support tickets, product usage data with related documentation, or sensor readings with maintenance reports.

AI models need the ability to see the full data picture to produce accurate results. By keeping structured and unstructured data together through our extensible platform, you're not just moving data, you're preserving the rich context that makes AI applications truly intelligent.

Access the full details and start building real data context.

Higher Speed and Lower Costs with Direct Loading

We’re always listening to our customers as we make Airbyte a more complete data movement platform. And we heard you loud and clear: warehouse destination bills were going up.

Airbyte's new Direct Loading feature replaces the intermediate step of storing raw JSON data in your warehouse, now instead writing typed data directly to final tables. This eliminates the problem of ever-growing raw tables and reduces compute costs by handling data type conversions before the data reaches your warehouse. The end result is the same clean data you expect, but with less storage and lower costs. Depending on the use case, projected warehouse compute cost savings can be between 50-70%.

Direct Loading is now available for BigQuery. ClickHouse and Snowflake are coming soon, with more destinations being added continuously. Read this and learn how to start using direct loading.

Try Upgraded Airbyte Today

Whether you are new to Airbyte or a long-term user, we’re excited for you to try out our new features. Start a free trial or contact our team to learn how these new capabilities can transform your data infrastructure.

Limitless data movement with free Alpha and Beta connectors

Introducing: our Free Connector Program ->

The data movement infrastructure for the modern data teams.

Try a 30-day free trial