How Social Media Algorithms Shape What We See
The average person spends 2 hours and 21 minutes per day on social media. From eye-catching photos, viral videos, breaking news, and thought-provoking articles, scrolling through social media can feel like an rabbit hole of digital content. Behind these curated social media feeds are AI-powered ranking and recommendation systems that use individually tailored information. To assure that organizations are reliably developing and using AI, international standards, such as ISO/IEC 5338, ISO/IEC 42001, and ISO/IEC 23053, provide frameworks that support responsible AI management and systematic evaluation.
What Is a Social Media Algorithm?
Algorithms in social media platforms refer to a set of computational rules, data, and automated signals used to sort, rank, and deliver content in a personalized feed instead of showing posts in strict chronological order. They can be, for instance, content-based (personalized to a user’s unique taste) or based on specific posts that the system guesses the user will like.
Coders often write algorithms that make use of machine learning: algorithms “learn” how to carry out tasks under various levels of human oversight. Algorithms manage tasks like content filtering, feed ranking, and user-behavior prediction that would be tedious for humans to carry out. Ultimately, their main goal is to keep users on a platform as long as possible, so the platform can show them more ads.
Social Media Ad Management
AI benefits social media platforms by optimizing ad management. In other words, AI-powered tools help people who work in advertising, sales, and marketing analyze hundreds or thousands of ad targeting and budget variations, find and segment audiences, test ads, and improve speed and performance of ads in real time to get the best results.
How Do Algorithms in Social Media Work?
Every social media platform has its own flow of algorithms that tend to be heavily based on machine learning and ranking signals. Ranking signals refer to specific data points, user behaviors, and content attributes that platform algorithms use to decide what order posts, reels, or stories appear in a user’s feed. With each swipe, algorithms enable users to encounter a constantly changing mix of information, entertainment, and opinions that are curated to capture their attention and reflect their interests. This curated social media feed is due to the use of AI and machine learning techniques.
Although these AI and machine learning capabilities can make social media more personalized and easier to navigate, they also raise questions about transparency, privacy, bias, and the potential spread of misinformation. As these systems evolve, it is increasingly important to implement standards that assure AI is developed and managed responsibly.
International Standards for AI
ISO/IEC standards provide frameworks and guidance for responsibly developing and managing AI systems.
ISO/IEC 5338: AI System Life Cycle Processes
ISO/IEC 5338 defines a set of processes and associated concepts for describing the life cycle of AI systems (from development through operation and improvement) based on machine learning and heuristic systems.
ISO/IEC 42001: Artificial Intelligence Management System (AIMS)
ISO/IEC 42001 specifies the requirements and provides guidance for establishing, implementing, maintaining and continually improving an AIMS within the context of an organization. It helps organizations establish policies, responsibilities, and processes for managing AI-related risks and opportunities.
ISO/IEC 23053: Framework for AI Systems Using Machine Learning
ISO/IEC 23053 establishes an artificial intelligence (AI) and machine learning (ML) framework for describing a generic AI system using ML technology. The framework describes the system components and their functions in the AI ecosystem.
Where to Find ISO/IEC 5338, ISO/IEC 42001, and ISO/IEC 23053
Besides being available on the ANSI Webstore, these ISO/IEC standards are also available as part of these Standards Packages:
- ISO/IEC 22989 / ISO/IEC 23053 – Artificial Intelligence Set
- ISO/IEC 27001 / ISO/IEC 42001 – Artificial Intelligence Set
- ISO/IEC 42001 / ISO/IEC 38507 / ISO/IEC 23894 – Artificial Intelligence Risk and Governance Package
- ISO/IEC 42001 / ISO/IEC 42005 / ISO/IEC 42006 – Artificial Intelligence Package
- ISO/IEC 42001 / ISO/IEC 22989 / ISO/IEC 23894 – Artificial Intelligence Package
- ISO/IEC 42001 / ISO/IEC 23894 – Artificial Intelligence Set
- ISO/IEC 42001 / ISO/IEC 23894 / ISO/IEC 42006 – Artificial Intelligence Package
- ISO/IEC 42001 / ISO/IEC 27001 / ISO/IEC 27002 – Artificial Intelligence and Security Management Package
- ISO/IEC 42001 / ISO/IEC 42005 – AI Management and Impact Assessment Set
- ISO/IEC 5338 / ISO/IEC 8183 / ISO/IEC 42001 – Artificial Intelligence Package
