[FederatedLearning][Post-v1] Full on-device confidential training in iOS/Android TEEs

[ooples]

Out of scope for FL v1.\n\nDetails: docs/FederatedLearning_OutOfScope_Backlog.md\n\nGoal: Investigate feasibility and implementation of running training/inference inside iOS Secure Enclave / Android TrustZone rather than via an attested gateway runtime.

[coderabbitai]

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

✅ Successfully generated implementation plan.

💡 Use @coderabbitai <feedback> to replan

Examples

• Example 1
• Example 2

🔗 Similar Issues

Related Issues

• [Phase 3] Implement Federated Learning Framework #398
• [FederatedLearning][Post-v1] Create post-v1 backlog tracking issue #536

👤 Suggested Assignees

• ooples

---

🧪 Issue enrichment is currently in early access.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

[coderabbitai]

Warning

This ticket appears to be from a different project or based on architectural plans that were never implemented in this codebase. The ticket references "Out of scope for FL v1" and a backlog document (docs/FederatedLearning_OutOfScope_Backlog.md) that does not exist. The codebase has no federated learning v1 implementation and no "attested gateway runtime" architecture that this ticket suggests should be compared against. Additionally, the iOS Secure Enclave fundamentally cannot perform ML operations—it only supports P-256 elliptic curve cryptography, making half of the ticket's scope technically impossible on that platform.

You can reply directly to this comment with additional context or clarifications to help CodeRabbit complete the planning task, with an @-mention to the bot.