Snapchat joins other platforms in the fight against ‘AI slop’.

In a significant move that underscores a growing industry-wide concern, Snapchat has officially joined the ranks of major online platforms like YouTube, LinkedIn, and Substack in their concerted effort to combat the proliferation of "AI slop" – content, including text, images, and videos, that is entirely generated by artificial intelligence tools. This surge in synthetic media, fueled by the rapid advancement and increasing accessibility of generative AI, has led to an online landscape saturated with machine-created essays, hyper-realistic videos, and other forms of content that often lack originality and authenticity. Snap, the parent company of Snapchat, announced on Friday that it will no longer recommend "wholly AI-generated videos" within its highly popular Spotlight feed, signaling a clear preference for "authentic, human-made content." This strategic shift aligns Snapchat with the recent policy updates from YouTube, LinkedIn, and Substack, all of which are grappling with the challenges posed by AI-generated material.

It is important to note that Snapchat is not implementing a blanket ban on all AI-generated content. The platform itself offers AI tools for content alteration, and as such, AI-enhanced or edited content will continue to be featured in user recommendations. However, Snap’s decision reflects a candid acknowledgment of the inherent drawbacks of entirely AI-generated content, which is often characterized as "low-quality," "repetitive," and generally unappealing to its user base. This sentiment is echoed by recent research indicating a widespread public agreement with these descriptors. Furthermore, a separate survey highlights a concerning trend: the more exposure users have to AI-generated content in their social media feeds, the less likely they are to trust the authenticity of any content they encounter online.

Chris Best, co-founder and chief executive of Substack, eloquently articulated this growing unease last week, stating, "It’s getting harder to tell what’s real on the internet." In conjunction with the launch of a new tool designed to help readers identify AI-generated writing on the newsletter platform, Best cited research suggesting that up to 40% of content on social media is now either fake or AI-generated. He further warned that "Platforms that reward fakeness will create a race to the bottom," emphasizing the potential for such practices to degrade the overall quality of online discourse.

LinkedIn, a platform specifically curated for professional networking, has also taken decisive action. This week, it introduced a dedicated button that empowers users to report posts or comments they suspect are AI-generated. Hari Srinivasan, LinkedIn’s chief product officer, declared "AI slop is a top priority for all of us," revealing that in just the past couple of months, the platform has thwarted "billions" of attempts to post AI-generated comments. He elaborated that "Every day we are now catching hundreds of thousands of automated comment attempts," underscoring the sheer volume of such activity.

Mirroring Snap’s nuanced approach, LinkedIn is not entirely disavowing AI. Users who employ AI tools to "refine" their posts are not the target of these new measures. However, LinkedIn is phasing out an automated prompt that previously offered to "enhance" posts with AI assistance, reverting to a more traditional proofreading tool. This adjustment signifies a commitment to fostering genuine human expression rather than automated content generation.

Meanwhile, YouTube, the video-sharing behemoth owned by Google, has also updated its policies this month, focusing on the monetization of AI-generated content. Research from the previous year uncovered a significant number of YouTube channels that relied exclusively on AI-generated videos, many of which boasted millions of subscribers and generated substantial revenue. To curb the influx of what YouTube terms "inauthentic content," the platform has categorized three styles of videos as ineligible for monetization: generic, repetitive, or template-based. These categories are particularly susceptible to rapid creation using generative AI tools.

Matt Halprin, YouTube’s chief of trust and safety, explained in an interview regarding the updated rules that while AI tools can be beneficial for content creation, they are increasingly being exploited to produce deceptive and low-quality videos. He acknowledged the dual nature of the technology: "The same technology really enables great stuff," he stated, "But it also enables stuff that’s kind of content farming, and that’s the stuff that we don’t want to have." This distinction highlights YouTube’s objective to encourage creativity and value while deterring the mass production of unoriginal and potentially misleading content. The collective actions of Snapchat, YouTube, LinkedIn, and Substack signal a unified front against the unchecked spread of AI-generated content, aiming to preserve the integrity and authenticity of online information and user experiences. The ongoing evolution of AI technology necessitates continuous adaptation from these platforms to ensure a healthy and trustworthy digital environment for all.