Spotify Text Feedback Boosts Recs
Spotify’s upcoming update introduces text-based feedback for recommendations. It shifts from passive data tracking to active user guidance. Character limits ensure controlled influence.
Search
Tag: #recommendations18 results
Facebook Opt-In Camera Roll Suggestions Launch in EU/UK
Facebook is rolling out an opt-in feature for EU and UK users that surfaces relevant photos and videos from their camera roll. It also suggests fun edits and collages to enhance sharing. The feature respects regional privacy requirements by requiring user opt-in.
Spotify Expands Prompted Playlists to Podcasts
Spotify expanded its beta Prompted Playlists feature to podcasts on Tuesday, allowing Premium users to generate custom episode playlists via text prompts. Originally for music since December, it now helps discover new shows by steering the recommendation algorithm. Currently available only in English for US and Canada Premium users.
Apple iOS 26.5 Public Beta Released
Apple has released the iOS 26.5 public beta, featuring Suggested Places in Apple Maps that shows trending spots like restaurants based on location or search history. The update introduces location- and search-based ads in Maps, clearly marked with privacy protections ensuring data stays on-device. It also retests end-to-end encryption for RCS messages, with rollout uncertain.
iOS 26.5 Beta Sets Stage for Maps Ads
Apple released the first beta of iOS 26.5. The update lays groundwork for advertisements in Apple Maps. It also introduces the Suggested Places feature.
ChatGPT Nails Streaming Guide Across 6 Apps
A user prompted ChatGPT to recommend content across six streaming apps. It delivered surprisingly accurate suggestions, acting as an effective guide. This simple approach cuts through endless scrolling.
Edit Spotify Taste Profile for Rec Control
Spotify now allows users to edit their Taste Profile. Changes impact personalized playlists like Discover Weekly, general recommendations, and Wrapped. This gives more user control over AI-driven personalization.
LLM Agents Auto-Optimize RecSys Models
A self-evolving system uses Google's Gemini LLMs to autonomously generate, train, and deploy recommendation model improvements. It features an Offline Agent for hypothesis generation and an Online Agent for production validation. Deployed successfully at YouTube, surpassing manual workflows.