Meta, Instagram, and TikTok Hand Algorithm Control to Users—Here's How It Works
Social platforms are replacing opaque recommendation systems with user-facing tools that let people explicitly shape what they see in their feeds.
Last verified:
Social Media Platforms Shift From Opaque Feeds to User-Directed Recommendations
Social media companies have historically gatekept algorithmic ranking, forcing users to accept whatever the recommendation system surfaced. That model is reversing. According to TechCrunch AI, Meta’s Threads, Instagram, and TikTok have rolled out features between December 2025 and June 2026 that move feed curation from black-box algorithms to explicit user instruction. The shift trades platform unilateral control for measurable engagement gains and user satisfaction.
Threads’ Two-Tier Preference System
Meta’s Threads launched “Your Algo” on June 16, 2026, building on the platform’s earlier “Dear Algo” tool, which debuted in February. According to TechCrunch AI, “Dear Algo” required users to publish public posts (e.g., “Dear Algo, show me more posts about podcasts”) to signal content preferences. The new “Your Algo” feature privatizes this interaction: users can specify topics they want to see more or less of—such as requesting more baseball content and less stressful news—and set the duration of each preference to one, three, or seven days. The system adapts the feed in real time without broadcasting intent to followers.
Instagram Extends Topic Control Across All Feed Surfaces
Instagram launched “Your Algorithm” in early June 2026, expanding a feature that debuted in the reels feed in December 2025. According to TechCrunch AI, the tool now covers the main feed, explore tab, and reels. Users access the feature in settings to view topics Instagram’s model identifies as their interests, then explicitly customize recommendations by indicating what they want to see more or less of. The system adjusts recommendations accordingly. Instagram head Adam Mosseri attributed the shift to large language models, which TechCrunch AI reports can surface the reasoning behind content ranking and enable users to communicate preferences directly—a transparency improvement over historically opaque recommendation models.
TikTok’s Topic Management Approach
TikTok offers “Manage Topics,” a tool that gives users granular control over the “For You Page” (FYP), TechCrunch AI notes. The feature allows users to add or remove topics from their recommendations, similar to Instagram’s approach but integrated into TikTok’s short-form discovery surface.
Why This Matters
The shift from algorithmic opacity to user direction represents a fundamental restructuring of feed economics. Platforms retain engagement gains—users spend more time on feeds that match stated preferences—while users gain predictability and reduce algorithm-induced discovery friction. This model succeeds only if the underlying rankers can reliably incorporate user signals without degrading diversity or introducing gaming incentives (users might demand only positive content or engagement-bait topics). Teams evaluating social media’s role in content strategy should monitor whether these tools suppress reach for niche content or whether transparency encourages broader exploration. The long-term question is whether user-directed feeds sustain the serendipitous discovery that historically drove engagement, or whether explicit preference systems fragment audiences into self-reinforcing silos.
Frequently Asked Questions
Why are social media companies adding user-controlled algorithm features?
According to TechCrunch AI, platforms gain engagement by showing users content aligned with their stated preferences, while users benefit from feeds tailored to their interests instead of opaque, one-size-fits-all ranking systems.
Can users keep their algorithm preferences private?
Yes. Threads' new 'Your Algo' feature lets users privately specify content preferences for 1, 3, or 7 days, whereas the earlier 'Dear Algo' tool required public posts to influence recommendations.
How do large language models improve recommendation transparency?
Instagram head Adam Mosseri stated that LLMs enable recommendation systems to show users why content is displayed and allow explicit preference communication—improvements over historically opaque ranking models, according to TechCrunch AI.