The app for tracking TV, movies, podcasts, and everything
apple privacy
| Source: Mastodon | Original article
A new AI‑driven service called **Sofa** landed on the App Store this week, promising to become the single place users can log every episode, film, podcast and even audiobook they consume. The Verge’s preview shows a sleek interface that lets users type or speak natural‑language commands – “Add the latest season of *The Crown* to my watchlist” or “Remind me to finish *Serial* episode 5” – and the on‑device language model instantly updates a unified library.
Sofa distinguishes itself with a privacy‑first architecture: all metadata stays on the user’s device, and the LLM runs locally on Apple’s M‑series chips, eliminating the need to send listening habits to the cloud. The app also pulls schedule data from major broadcasters, integrates with Apple TV, Spotify and Audible, and can generate personalized recommendations based on the user’s own consumption patterns rather than a centralised profile.
Why it matters is twofold. First, it tackles the fragmentation that has long plagued media tracking – users juggle Trakt, Letterboxd, JustWatch and separate podcast apps, each with its own login and sync quirks. By unifying these feeds under a single, AI‑enhanced hub, Sofa could set a new standard for how we organise digital entertainment. Second, its on‑device LLM showcases the next generation of consumer privacy tools, echoing the capabilities we explored in our April 5 coverage of Google’s Gemma 4 models and their potential for local inference.
What to watch next: Sofa’s rollout is currently limited to iOS 17, with an Android beta slated for later this quarter. The developers have hinted at a tiered subscription that will unlock deeper analytics and cross‑device sync, while competitors may respond with their own AI‑powered add‑ons. Observers will also be keen to see whether Apple’s upcoming privacy enhancements in iOS 26 make on‑device LLMs a default feature for third‑party apps. If Sofa delivers on its promise, the way we catalogue our media lives could shift from scattered spreadsheets to a single, conversational companion.
Sources
Back to AIPULSEN