Snap Limits AI-Only Videos As Platforms Rework Feed Quality

Snap is drawing a sharper line between AI-assisted editing and fully synthetic feed content as consumer platforms try to protect trust in recommendations.

BBC reported that Snap said on Friday it would stop recommending wholly AI-generated videos in Snapchat's Spotlight feed and favor authentic, human-made content.

The policy does not remove AI from Snapchat.

Snap already offers tools that can enhance or edit user posts, so edited material can still appear in recommendations.

The change targets videos made entirely by generative systems, which Snap characterized as low-quality, repetitive and misaligned with what users want to see.

Feed Trust Becomes The Moderation Problem

The operating problem is not only whether a post was made with AI.

Generative tools have made synthetic text, images and videos easier to produce at scale, creating a moderation problem across platforms that rely on automated feeds.

If users repeatedly encounter fake clips or automated comments, ordinary posts can start to look less trustworthy even when they are genuine.

Substack, LinkedIn and YouTube have each moved in the same direction, but their remedies reflect different products.

Substack’s part of the platform response centers on AI-written newsletters and posts.

Co-founder Chris Best cited research finding that up to 40% of writing on social media is fake or AI-generated and warned that it was becoming harder to tell what is real online.

LinkedIn added a reporting button for posts or comments that appear AI-generated.

Chief product officer Hari Srinivasan put the platform's blocked attempts to post AI-generated comments at billions in recent months and wrote that it now catches hundreds of thousands of automated comment attempts every day.

LinkedIn is also removing an automated prompt that offered to enhance user posts with AI, returning that feature to simpler proofreading.

Monetization And Reach Are The Enforcement Levers

YouTube's approach shows why platform economics matter.

The Google-owned video platform updated monetization rules this month for videos it classifies as generic, repetitive or template-based, the formats that generative tools can mass-produce quickly.

Research found that some YouTube channels made solely of AI-generated content had millions of subscribers and, in some cases, millions of dollars in revenue.

The common thread is reach control rather than a blanket ban on AI tools.

LinkedIn said users who use AI to refine posts should not be caught by anti-slop measures, while Snap left room for AI-enhanced or edited content.

That distinction creates a practical enforcement test: platforms must identify when AI changes become content farming instead of ordinary assistance.

creators and users, the next measurable issue is whether demotion and reporting tools reduce low-quality synthetic material without suppressing legitimate editing workflows.

A survey cited in the story found that people exposed to more fake AI-generated content become less likely to believe online content is genuine, making feed integrity a trust metric rather than a narrow moderation feature.