Anthropic’s New AI Data Policy Explained: 30-Day Retention vs OpenAI’s Privacy Approach

Anthropic’s planned customer-managed storage keeps a 30-day retention window. Explore control, security monitoring, and the comparison with OpenAI.

This content is blocked because it would connect to YouTube.
This content is blocked because it would connect to Spotify.

Anthropic’s Enterprise Data Retention and Cloud Control

Anthropic is changing how enterprise AI data can be stored—and the decision could reshape the privacy and security battle with OpenAI.

Reuters reports that Anthropic plans to keep its 30-day data-retention requirement for businesses using powerful models such as Fable, Mythos and future frontier systems. The major change: enterprise customers would be able to store those retained records inside their own cloud infrastructure.

Why does that matter? Companies handling confidential, proprietary or regulated information want tighter control over:
• Who can access AI activity data
• Where information is geographically stored
• How records fit existing security policies
• Whether retention complies with industry regulations
• How quickly information can be audited or deleted

Anthropic says the 30-day window is intended to help detect cyberattacks and other dangerous misuse of its technology. The company has reportedly worked with more than 100 customers, including Salesforce, to design the new system.

But OpenAI is taking a different approach. Reuters notes that OpenAI announced a safety system that it says can identify potential misuse without retaining customer data. That creates a defining enterprise-AI debate: should providers temporarily retain activity for security monitoring, or minimize collection while attempting to detect abuse another way?

This Daily AI Chat Short explains:
• Anthropic’s planned enterprise data-retention change
• Why customer-managed cloud storage improves control
• The security case for a 30-day monitoring window
• The privacy and compliance concerns facing businesses
• How Anthropic’s strategy differs from OpenAI’s
• Why data governance could determine enterprise AI adoption

For CIOs, cybersecurity leaders, privacy professionals and business technology teams, model intelligence is only part of the decision. Trust, auditability, data sovereignty and control are increasingly important when choosing an AI platform.

Source: Reuters, August 20, 2026.
Reporting by Deborah Sophia in Bengaluru.
Editing by Diti Pujara.

What to watch

Where records are stored and how long they are retained are separate decisions. Customer-managed infrastructure may improve control while leaving retention obligations in place. Businesses still need to understand access, auditing, deletion, and how provider claims fit their own requirements.

Related reading: our coverage of AI-agent safeguards and OpenAI’s training pause.

Watch and listen

Watch the YouTube Short above or listen to the full episode on Spotify.

Scroll to Top