Rubrik brings BigQuery backup to general availability, extends cover to Cloud Storage

Rubrik has made agentless, immutable backup for Google BigQuery generally available as of November 11, 2026, and promised protection for Google Cloud Storage in the coming months. The company is pitching the expansion at organizations whose analytics and AI workloads increasingly depend on cloud data that native recovery tools cannot protect on their own.

The half-launched state of the announcement matters. BigQuery protection is real and shipping today; Google Cloud Storage protection is a roadmap commitment without a date. Rubrik's headline conflates the two, and buyers should read the fine print before assuming their object storage is covered.

Why does BigQuery backup matter now?

Rubrik's argument rests on a shift in how enterprises use BigQuery. The platform has long served as a data warehouse (a centralized repository for reporting and analysis), but it is increasingly becoming the layer that AI agents query and act on directly. That raises the stakes if data is deleted, corrupted, or encrypted in an attack.

Rubrik backed the concern with figures from its research arm, Rubrik Zero Labs. The company said 90% of global IT and security executives reported cyberattacks, and that cloud or software-as-a-service environments accounted for 28% of those breaches. The most striking number: 74% of respondents reported that threat actors had successfully compromised backup and recovery systems.

That last figure is doing heavy lifting in the announcement. It comes from Rubrik's own research operation, which has a commercial interest in documenting backup insecurity. Rubrik did not publish the survey methodology or sample size alongside the launch. The numbers are directionally consistent with broader industry reporting, but readers should treat vendor-cited statistics about vendor-addressed problems with appropriate caution.

Anneka Gupta, Chief Product Officer at Rubrik, framed the stakes in terms of AI trust rather than traditional recovery. "The question for every AI deployment is no longer what the agent can do, but what happens when its data is gone. If that data can be lost or corrupted, the agents can't be trusted," Gupta said.

What is actually shipping?

The BigQuery product is available now as part of Rubrik's wider Google Cloud data protection portfolio. Its technical claims are concrete, though several rest on how the service behaves in production rather than what the announcement describes.

• Restore granularity. Customers can restore an entire dataset or a single table while keeping views and routines intact.
• Scale and flexibility. Recovery works at petabyte scale, with the option to recover data in place or export it to another project or region.
• Storage choice. Backups can live in a customer's own Google Cloud Storage bucket or in Rubrik Cloud Vault, using Google Cloud Storage Object Retention Locks to enforce immutability (backup copies that cannot be altered or deleted during a set retention period).
• No agents. The service requires no agents or separate compute deployment. It uses incremental backups with periodic full backups, a model Rubrik says reduces operational overhead.
• Retention beyond native limits. Customers can keep BigQuery backups beyond the platform's native seven-day time travel limit, with unified compliance reporting intended to create an auditable recovery record for regulated sectors.
• Unified management. BigQuery protection runs alongside the rest of a customer's Google Cloud estate under the same service-level agreement policies and role-based access controls.

What the announcement does not say: pricing. Rubrik did not disclose cost for the BigQuery product, and independent performance benchmarks for petabyte-scale restores under the new architecture are not yet available. Both questions will determine whether the offering is practical beyond early adopters.

"Rubrik now makes BigQuery recoverable and extends that protection to Google Cloud Storage, so customers can scale AI knowing they can recover from deletion, corruption or an attack," Gupta said in the second of two statements. The second half of that sentence describes functionality that, by Rubrik's own timeline, does not fully exist yet.

Is native cloud recovery really not enough?

The broader industry argument Rubrik is joining holds that organizations need recovery capability independent of the production environment. Native cloud recovery features can provide short-term rollback, but security vendors have argued these tools fall short when attackers gain privileged access or tamper with backup settings.

That argument has gained traction as cyber incidents have shifted from endpoint compromise toward attacks on identity systems, management planes, and backup infrastructure itself. In cloud environments, the risk is less about hardware failure and more about account compromise, configuration changes, and deliberate deletion.

The threat model is credible. The proposed remedy still requires trust in the backup vendor's own immutability guarantees, which Rubrik grounds in Google's Object Retention Locks rather than proprietary technology. That dependency on the underlying cloud provider cuts both ways: it anchors immutability in a native control, but it also means protection inherits whatever risks attach to the customer's Google Cloud account configuration.

For companies using BigQuery to support analytics, reporting, and machine learning, a data loss incident would not stay inside IT operations. Rubrik notes that if downstream applications or AI systems rely on the data, disruption can affect decision-making, customer services, and internal workflows.

Where does this fit in the market?

The update lands in a growing market for cloud-native data protection, as backup suppliers adapt traditional recovery products for managed cloud services. Protecting object storage and large-scale cloud data platforms has become more prominent as enterprises move critical workloads into hyperscale environments.

Google Cloud Storage serves as a foundational storage layer for enterprise applications, archives, and analytics pipelines. BigQuery sits at the center of many data estates because it handles reporting, data engineering, and AI-related processing from a single environment. Covering both makes commercial sense for Rubrik; covering one today makes the announcement half-finished.

The competitive context is worth watching. Google offers its own native protection for these services, and other backup vendors are racing to add managed-cloud coverage. Rubrik's differentiation rests on isolation from the production environment and compliance-grade reporting, not on exclusivity.

The practical test arrives when Google Cloud Storage protection actually ships. Until then, Rubrik has delivered a solid BigQuery recovery product, a credible AI-era threat narrative, and a roadmap claim that customers will need to hold the company to.