Apple rebuilds search infrastructure powering Spotlight and Mail
Apple announced at WWDC 2026 that it rebuilt the search infrastructure powering Spotlight, Photos, and Mail across its next-generation OS platforms, and introduced a developer API for on-device, LLM-powered semantic search. The company says the new index is more stable and comprehensive, reindexes existing content, and makes new content searchable almost immediately after an update. For developers, the more significant change is SpotlightSearchTool paired with LanguageModelSession, which lets apps build retrieval-augmented search experiences using Apple's on-device Core Spotlight framework, with custom pipeline stages for tasks like sentiment analysis. Early macOS 27 beta testers report faster indexing, though some note Spotlight's basic keyword search has long been unreliable, so results may vary as the rebuild rolls out.
Apple's search-infrastructure rebuild is less about faster file lookups and more about giving developers a sanctioned way to build retrieval-augmented, on-device semantic search into their apps - a meaningful expansion of Apple's on-device AI developer toolkit that arrived alongside the company's broader Apple Intelligence and Siri overhaul at WWDC 2026.
What happened
Apple announced during its WWDC 2026 keynote that it has rebuilt the search infrastructure powering Spotlight, Photos, and Mail across its next-generation software platforms (iOS 27, iPadOS 27, and macOS 27 "Golden Gate"), according to MacRumors. Apple says the index has been rearchitected to be more stable, efficient, and comprehensive, and that after updating, the new infrastructure reindexes existing device content automatically, with new content becoming searchable "almost immediately." In a companion WWDC developer session, Apple detailed a new Core Spotlight-based toolkit - including SpotlightSearchTool and LanguageModelSession - that lets developers "level up basic search into a retrieval-augmented system," with delegate-based hydration patterns, custom PipelineStages for tasks such as sentiment analysis, and guidance on how metadata quality affects search results.
Technical context
Apple's own materials frame this as layering semantic search on top of the existing keyword-based Spotlight index. Per independent analysis from longtime Apple commentator Howard Oakley, semantic search works more like a thesaurus than a strict Boolean index, matching intent rather than exact keywords, and is powered by per-app Core Spotlight indexes stored locally (in ~/Library/Metadata rather than the system-wide .Spotlight-V100 volume index). That keeps semantic search opt-in per app and preserves Apple's on-device privacy model rather than centralizing everything into one global index.
For practitioners
For ML and app engineers, the practical takeaway is that Apple is offering a first-party path to ship RAG-style search inside apps without standing up a separate vector database or embedding pipeline - LanguageModelSession plus Core Spotlight indexing handles the retrieval step on-device. That lowers the barrier to adding natural-language search to iOS and macOS apps, but it also ties the feature to Apple's on-device model stack and Core Spotlight's per-app indexing rules, so teams should evaluate how metadata quality and pipeline-stage design affect retrieval accuracy before relying on it for production search.
What to watch
Real-world beta feedback is mixed so far: some macOS 27 testers report Spotlight's basic search is noticeably faster after the rebuild, while other developers and users note that simple keyword search has been unreliable for years and question whether the new infrastructure - rather than the added semantic layer - actually fixes that. Practitioners building on the new APIs should watch for documentation updates, further WWDC session follow-ups, and how metadata-quality guidance evolves as iOS 27 and macOS 27 near their fall release.
Key Points
- 1Apple rebuilt the search infrastructure behind Spotlight, Photos, and Mail, with faster indexing after the macOS 27 and iOS 27 updates.
- 2A new developer toolkit pairs SpotlightSearchTool with LanguageModelSession to build on-device retrieval-augmented semantic search inside iOS and macOS apps.
- 3Semantic search runs through per-app Core Spotlight indexes stored locally, keeping Apple's on-device privacy model intact while adding intent-based matching.
Scoring Rationale
Beyond a search-infrastructure refresh, Apple's WWDC materials reveal a new on-device retrieval-augmented search toolkit (SpotlightSearchTool plus LanguageModelSession) for iOS/macOS developers, directly relevant to ML and app engineers building semantic search features. It is a meaningful platform capability rather than a frontier-model or industry-shaking event, so it sits in the notable range.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
