Unbubble Glance

Glance checks a variety of signals from the post and it's replies, looking for contrary opinions and clearly manipualtive language. When in doubt, it searches the web to get more context and cross-check the claims. If it finds strong contradictory opinions or clear signs of sensationalism, it flags the post for you to take a closer look.

Yes, Glance uses LLMs, along with other more traditional technologies, to understand the content of the posts and cross-check it against other sources. It only reads the content of public posts, and none of it is used for training or improving the models.

X is where fast-moving, low-context posts spread the fastest, so it's where a tool like Glance has the most impact. Reddit, Instagram, and Facebook are on our roadmap.

Yes. Glance is free and basic usage will always stay free. We may introduce limits and a paid plan for heavy users down the road, to keep the service sustainable and to prevent abuse, but the core experience stays free.