Instagram adds new feature letting users personalize their feed algorithm

2026-06-11 · Source: Dataconomy · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Emerging Technologies & Innovation · Depth: Novice, quick

Summary

Instagram has introduced a new feature called "Instagram, Your Algorithm" that allows users to personalize their main feed by selecting topics they wish to see more or less of in their recommendations. Adam Mosseri, head of Instagram, stated this change aims to give users greater control over their app experience, believing that providing meaningful control benefits the business. Mosseri acknowledged that while algorithmic recommendations are a technical achievement, they impose a "cost" on individual agency, as users previously could not explicitly tell the system their preferences. The new personalization features are enabled by Large Language Models (LLMs), which make previously "unreadable" ranking algorithms understandable by analyzing content clusters and describing them in user-friendly terms, facilitating deeper future personalization.

Key takeaway

For AI Product Managers designing user-facing recommendation systems, Instagram's "Your Algorithm" feature demonstrates that empowering users with direct control over algorithmic inputs can enhance satisfaction and drive business value. You should explore how Large Language Models can translate complex ranking logic into understandable user choices, fostering greater agency and personalization within your products. This approach mitigates the "cost" of opaque algorithms.

Key insights

Large Language Models empower users to directly influence algorithmic content feeds by translating complex recommendations into understandable, adjustable topics.

Principles

• Meaningful user control benefits business.
• Algorithmic opacity reduces user agency.
• LLMs can demystify complex ranking models.

Method

LLMs analyze content clusters to describe them in user-understandable terms, allowing users to select topics for increased or decreased visibility in their main feed.

In practice

• Adjust feed topics directly.
• Influence content recommendations.
• Tailor app experience.

Topics

• Instagram
• Algorithmic Personalization
• Large Language Models
• User Control
• Feed Customization
• Recommendation Systems

Best for: Product Manager, AI Product Manager, Director of AI/ML, AI Ethicist

Related on AIssential

• Hands on with X’s new AI-powered custom feeds — Grok's AI powers X's new custom timelines by semantically labeling posts, moving beyond keyword-based feeds.
• PSA: If you use the Meta AI app, your friends will find out and it will be embarrassing — Meta's aggressive AI app promotion and interconnected data sharing raise significant user privacy and adoption challenges.
• Why AI Chatbots Are the Future of Personalized Marketing — AI chatbots leverage real-time learning, predictive analytics, and generative AI for scalable, personalized customer engagement.
• Beyond the Scroll: How Social Media Algorithms Shape Your Reality — Social media algorithms, driven by engagement and recency, rapidly create echo chambers by exploiting human negativity bias.
• How the “dear algo” meme forced Meta to change its feed forever — Meta's "dear algo" feature allows users to directly modify their Threads feed recommendations in real time.
• Bond, a new social media platform, wants to use AI to help you kick your doomscrolling habit — Bond uses AI to transform user-posted memories into personalized real-world activity recommendations.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Dataconomy.