Built-in AI in the wild: A Mastodon translation success story

Описание: Mastodon posts mislabeled with the wrong language as well as costly translations that depend on a server: Those are common problems for many Mastodon instances. In this talk, Thomas Steiner from Chrome's Built-in AI team details how he integrated the Language Detector and Translator APIs into the popular Mastodon client Elk.zone. This isn't a textbook success story—you'll hear about what initially went wrong, the technical hurdles encountered, and how they were overcome to build a feature that now runs reliably in production.

Resources:
add support for the Language Detector API → https://goo.gle/47vgzNt
add support for the Translation API→ https://goo.gle/3WFV0U9
Elk Zone→ https://goo.gle/49ERpx1
Chrome built-in AI docs → https://goo.gle/3JHSWIi
Join the early preview program → https://goo.gle/49GPFU6


See more Web AI talks → https://goo.gle/web-ai
Subscribe to Chrome for Developers → https://goo.gle/ChromeDevs


Event: Web AI Summit 2025

Speaker: Thomas Steiner
Products Mentioned: AI for the web

#ChromeforDevelopers #WebAI