LinkedIn implements new measures to curb AI-generated slop

Read original on The Next Web (TNW)

#content-strategy#llm-moderation#social-medialinkedinlinkedin

Learn how major platforms are fighting back against AI-generated content saturation.

30-Second TL;DR

What Changed

Targeting AI-generated content that lacks human insight

Why It Matters

This shift forces content creators and marketers to rethink AI-assisted workflows, moving away from mass-produced content toward more nuanced, human-centric strategies.

What To Do Next

Audit your content automation pipeline to ensure your posts include unique, human-verified insights rather than generic LLM outputs.

Who should care:Marketers & Content Teams

Key Points

•Targeting AI-generated content that lacks human insight
•Algorithm adjustments to favor high-quality, authentic user interactions
•Effort to improve overall feed readability and user experience

Deep Insight

Background and context from public sources — not the original article. 22 sources cited.

Enhanced Key Takeaways

•LinkedIn is deploying "AI solving AI" systems to specifically identify and suppress generic posts, bot comments, and "attention-bait videos" rather than outright removing them, thereby reducing their distribution beyond immediate networks.
•The platform's new core algorithm, named "360Brew," utilizes a 150-billion-parameter AI model that evaluates content based on its substance, author credibility, and "dwell time" (how long users spend on a post), moving beyond traditional engagement metrics like likes.
•LinkedIn's measures also directly target and penalize "engagement pods" and automated commenting tools, which were previously used to artificially boost post visibility and engagement.
•This initiative comes despite LinkedIn itself offering built-in generative AI tools for users to enhance profiles and draft content, highlighting a distinction between AI assistance for quality and AI replacement for insight.

Technical Deep Dive

LinkedIn is employing "AI solving AI" systems, which are newly built technical systems trained with an in-house editorial team.
These systems are designed to differentiate content that offers original perspective, context, or expertise from that which is generic, repetitive, or lacks substance.
Classifiers are being developed to identify low-quality AI comments by analyzing language patterns and the volume/frequency of posts.
The new algorithm, referred to as "360Brew," is an AI system with 150 billion parameters that evaluates the actual text of posts, not just user reactions.
It assesses content substance, formatting signals, and author credibility as part of its quality classifier, which operates within minutes of publication.
Early tests of the new systems showed a 94% accuracy rate in correctly tagging generic content.
The system aims to learn over time by analyzing user engagement patterns and identifying language that adds perspective versus regurgitating existing ideas.

Future ImplicationsAI analysis grounded in cited sources

Authentic content will become a more critical differentiator for professionals.

The algorithmic shift to prioritize depth, expertise, and 'dwell time' over superficial engagement will compel creators to produce more substantive, original content to gain visibility and credibility.

An ongoing 'arms race' between AI content generation and detection will intensify.

As LinkedIn's detection methods improve, AI content generation tools will likely evolve to bypass these measures, creating a continuous cycle of refinement for both content creators and platform moderators.

Other professional or niche social platforms may adopt similar stringent content quality measures.

The success of LinkedIn's initiative in restoring content quality and user trust could set a precedent for other platforms grappling with the proliferation of low-value AI-generated content.

Timeline

2003-05

LinkedIn launched as a professional social networking platform.

2010s Early

LinkedIn began incorporating AI into its platform, alongside other major social media networks.

2022-12

The release of ChatGPT led to a 189% surge in AI usage in LinkedIn posts.

2025-01

Median post impressions on LinkedIn dropped by 13% in Q1 2025, reflecting a shift towards prioritizing higher-quality content.

2025-11-03

LinkedIn updated its terms to use member data by default for training generative AI models, with an opt-out option.

2026-01

LinkedIn began rolling out its 2026 algorithm changes, including the "360Brew" AI system, to penalize engagement bait and prioritize valuable content.

2003-05
LinkedIn launched as a professional social networking platform.
2010s Early
LinkedIn began incorporating AI into its platform, alongside other major social media networks.
2022-12
The release of ChatGPT led to a 189% surge in AI usage in LinkedIn posts.
2025-01
Median post impressions on LinkedIn dropped by 13% in Q1 2025, reflecting a shift towards prioritizing higher-quality content.
2025-11-03
LinkedIn updated its terms to use member data by default for training generative AI models, with an opt-out option.
2026-01
LinkedIn began rolling out its 2026 algorithm changes, including the "360Brew" AI system, to penalize engagement bait and prioritize valuable content.