Firefox

What? Neowin is pretty well known, that's weird, maybe a false positive?

Anyways, it was a shortish article so I'll just copy paste:

A few weeks ago, Mozilla announced that its Pocket and Fakespot services were getting the axe as the company focuses more on Firefox. It is a complete shutdown. Pocket, the read-it-later service Mozilla bought in 2017, will stop working on July 8, 2025. You have until October 8 to get your saved articles out before they are deleted forever. Fakespot, which helped you spot garbage product reviews, is also being sunsetted.

But the house cleaning does not stop with those two. Neowin has spotted a shutdown notice dated June 26, 2025 for Deep Fake Detector, the Firefox extension that was supposed to tell you if a piece of text was written by a human or an AI chatbot. That tool used a combination of Mozilla's own proprietary ApolloDFT engine and open-source models like ZipPy to give you a verdict on what you were reading.

The notice says:

** Important Update: Deepfake Detector will shut down on June 26, 2025. **

On June 26, you will no longer be able to use Deepfake Detector. Thanks for supporting our journey.

This brings us to the AI tools. Following the pattern, the Orbit website was updated with a banner that announced the service would shut down by June 26. Orbit was Mozilla's big privacy-first experiment in AI. It was a Firefox add-on that could summarize articles and answer questions about a webpage's content without sending your data to a third party.

Orbit's private, self-contained setup can be replaced with the new sidebar built directly into Firefox, letting you connect to third-party chatbots like ChatGPT and Gemini. But, for Orbit users, this is still a huge loss, as a key feature of the service was privacy. Your prompts were handled by Mistral LLM (Mistral 7B) within Mozilla's GCP instance and were not shared with other companies for model training.

Mozilla keeps saying these cuts are necessary. As the only major browser not owned by a tech giant, its resources are limited; hence, the need to focus its cash and engineering talent on the core Firefox browser to compete.