On 2 Oct 2026 Apple posted on Developer News that macOS Full Disk Access will get additional controls so grants require very explicit user action. The post names AI agents as the reason risk will grow substantially. No OS version or ship date was announced.

Key Takeaways

  • ✓Official: developer.apple.com/news/?id=p6zjojqw dated 2026-10-02
  • ✓Apple names AI agents as the reason FDA risk will grow substantially; FDA historically for backup-class apps
  • ✓Direction: grants only via very explicit user action; no API, UI, or ship date yet
  • ✓Press context (not Apple): TechCrunch ties the post to Muse desktop-access controversy and ChatGPT Mac reporting
  • ✓Dev action: audit FDA use in coding agents; prefer user-picked folders and security-scoped bookmarks
🧭

Heavy Claude Code use: compare subscription limits and API bills

Compare 29+ dev plans & simulate token costs vs $20/mo subscriptions

🔬

In-Depth Technical Analysis

Background

macOS Full Disk Access historically lets backup-class apps bypass normal privacy partitions. Desktop AI agents increasingly treat FDA as a one-click path to repos, mail, messages, and browser history. Apple’s 2 Oct 2026 Developer News post says some developers use FDA without users’ full understanding, and that risks will grow substantially as agents become more capable and autonomous.

Mechanism

Apple is not banning FDA. It will add controls so extraordinary access requires very explicit user action. No UI, API, or OS version was named. Prefer user-selected folders and security-scoped bookmarks per Apple’s sandbox file-access docs.

Benchmarks

None published. Treat this as policy intent until a beta ships. Press coverage linking Muse or ChatGPT Mac incidents is third-party context, not Apple’s attribution.

Developer playbook

Audit FDA prompts in coding agents; default to project-scoped access; update MDM/PPPC notes; watch the official Apple post for concrete ship details.

Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.