Facebook Adds 'Your Algorithm' AI Feed Control for Users

What happened

Meta has announced a new AI-powered feature called "Your Algorithm", which gives Facebook users more control over the algorithm that shapes what appears in their feeds. The feature was first launched on Instagram.

TechCrunch reported the update on 13 October 2026, framing it as a move to give people a more direct say in how their feed is ranked. The material available for this briefing covers the announcement itself. It does not set out the detailed mechanics of the controls, the rollout timeline or regional availability, so those points should be checked against Meta's own communications before being relied on.

Why it matters

Ranking systems have long been among the least visible parts of a social platform's service. Users experience the output of a recommendation model without much insight into how it was produced or how to influence it. Offering an AI-driven control over that ranking changes the relationship: the algorithm becomes a configurable part of the product rather than a fixed backdrop to it. For leaders in digital services, this is a reminder that AI personalisation is now a design surface that users can be invited to shape, not only a back-end optimisation.

The move also raises a broader question for any organisation deploying recommendation or prioritisation AI, whether in retail, public services or financial products. Once people are given agency over an automated decision, their expectations of transparency and predictability rise. Features of this kind will be judged less on their technical sophistication than on whether they make the system feel understandable and fair.

The Renascence take

Giving users a dial on an algorithm is a sensible response to declining trust in automated feeds, but a control only builds trust if people use it and can see what it does. Most readers will focus on the headline promise of control and miss the design question underneath: whether the setting is discoverable, whether its effect is visible, and whether it stays where users put it.

Control that people never find, or that they cannot see working, is simply a more elaborate default. Behavioural economics tells us that defaults usually win, so the real measure of "Your Algorithm" will be how many people change their settings and whether they notice a difference afterwards. Operators building similar features should treat the feedback loop as the product: show users what changed in their feed, and make the choice reversible in one step. If your algorithm is hard to explain, adding a switch will not make it trustworthy.