Netflix is putting generative AI directly into the screen where most of its members choose what to watch. On November 3, 2026 the company introduced AI Discover, a new tab in its TV app that it describes as a test kitchen for experimental ways of finding shows and films. The first experiment inside the tab starts from what a viewer is in the mood for rather than from a grid of rows.
The mechanics are simple. Viewers pick from a set of themes, each previewed with a clip from a title. Some describe a feeling, such as 'Something Light' or 'Scare Me'; others describe the company, like 'Family Movie Night', or the time available, like 'Quick Watches', and some reflect what is new, popular or recently watched. Choosing a theme brings up matching recommendations, narrower themes appear for those who want to keep digging, and a 'Something Different' option resets the set. Netflix says AI helps create the themes and that the app will explain why results appear.
Each title in the tab also carries a 'Tune Your Recommendations' button next to Play and My List. Viewers can use it to keep a suggestion or to tell Netflix not to recommend something again, and Netflix says AI Discover will remember that choice so the title stops resurfacing there. The company says the feature runs on a new intelligence layer that reacts to what members tell it in real time and keeps learning, but it has not described the models involved or how this layer relates to its long-standing recommendation systems.
The rollout is narrow. AI Discover is starting in the United States for members who use Netflix in English on supported TVs, with access widening gradually over the coming weeks. Members who have it will see a 'Discover' entry with a red gem icon and a 'New' label in the top navigation. Netflix frames the tab as a testing ground: member use and feedback will decide which experiments are kept, improved or dropped, and some may eventually move into the main Netflix experience.
Independent coverage matches the company's account. TechCrunch, which first flagged the launch, described the same themes, feedback controls and US English TV rollout, and The Verge separately reported the tab, the theme-based recommendations and the gradual US rollout. The Verge also noted that the launch follows reports of Netflix cutting around 1,000 jobs as growth slowed and viewing reportedly declined, context the company did not address in its post.
For the streaming industry, the test is a sign that the largest services see conversational, intent-driven discovery as the next step after algorithmic rows. Netflix has long relied on personalization to keep members engaged with a huge catalog, and starting from a mood or a time limit is a different interaction model from scrolling. It is also a cautious one: the feature lives in a separate tab, on TVs only, for a limited group, with no stated timeline for wider availability.
What remains unknown is whether AI Discover changes viewing behavior. Netflix has not published engagement data, has not said which AI systems generate the themes, and has given no detail on how feedback in the tab is used elsewhere in the service.