The keynote just wrapped up for move(data), our annual data + ai conference where Michel, the CEO and co-founder of Airbyte, laid out a three step plan for winning in AI. During the keynote, he also announced key features available in the latest release of the Airbyte Platform: Winter 2025. This release is intended to provide critical features required for you to use your first party data in AI tools and turn them into your competitive advantage.
To recap, too many organizations are hitting an AI wall, where they can’t reliably and safely share first party data with AI models. But, not doing so puts them at a disadvantage to their competitors. The three step plans solves this:
Bring all of your data together and clean it of sensitive data : Unlock structured and unstructured data from anywhere, and clean it before moving to an AI-ready data lake or destination to ensure data privacy. Keep your data under your control : Choose where you want to run your data pipelines: in the cloud, on-premises or via a VPC Maintain data portability : AI teams and data teams everywhere are converging on data lakes, because lakes offer unmatched flexibility to support multiple use cases—whether it’s AI, analytics, or operations. And, with AI moving so fast, you want to be prepared for what’s next.What’s in the Winter 2025 Release? Winter 2025 continues our focus on performance improvements, and adds the following new features to further accelerate AI adoption:
Data Access Enterprise Connector Bundle that offers new premium connectors for Oracle, SAP HANA, NetSuite, Workday, and ServiceNow, streamlining complex enterprise data integrationEnhanced GraphQL & OAuth 2.0 Support – Simplifies secure authentication and improves API-based data movement with more flexible and efficient integrations.Expanded File Transfer support to allow you to access all of your first-party data, including structured and unstructured data, along with metadata and permissions. Winter 2025 includes new support for Google Drive, SharePoint, and OneDrive (PDFs, videos, images). You can read a deeper dive here .
Data Control & Governance Support for AWS PrivateLink designed to deliver secure, private cloud-to-cloud data transfers, ensuring compliance and reducing exposure to public internet traffic. Enhanced Audit Logging – Detailed tracking of data movement activity to improve compliance monitoring and governance.Mappings for Data Privacy & Governance controls for built-in transformations to hash, encrypt, rename fields, and filter rows for GDPR, HIPAA, and SOC 2 compliance. Mapper operate at the source to clean data before ever hitting public AI models. Read a deeper dive blog here .
Data Portability & AI Workloads Apache Iceberg destination designed to aggregate your data and sync it into Iceberg to for highly scalable and performant AI and analytics workloads. And, with Iceberg’s schema evolution support, along with expanded files transfers, you can move unstructured data into Iceberg with confidence knowing your data is AI ready.
Platform & Performance Improvements Performance Upgrades for Syncs – speed increase across major connectors, including Amazon Ads, Google Sheets, Airtable, LinkedIn Ads, and Stripe. New Python-Based CDK – Enables faster and more efficient custom connector development, reducing integration time.Connection Tags & Resource Management – Organize, prioritize syncs, and allocate computing resources to critical data pipelines.OpenTelemetry (OTEL) Metrics – Provides deep observability into sync health, API activity, and data volume movement for improved pipeline monitoring.New Developer Hackathon And, to give you an opportunity to build using Winter Release, we’ve partnered with MindsDB for our next Developer Hackathon. Check out the event page for details on how you can enter for a chance to win some of the $5,000 prize pool!
That’s it for move(2025). You can find a complete list of sessions in the Youtube Playlist , and for more information on Winter 2025 check out the release notes and release home page.
