I don’t think AI is going to make disinformation worse.

Will Artificial Intelligence Really Worsen the Disinformation Crisis? A Nuanced Perspective

In recent discussions about the impact of AI on information integrity, a common concern has emerged: that artificial intelligence will dramatically escalate the proliferation of false or misleading content. Many fear that AI-generated “junk” will flood social media platforms at scale, making it more challenging for users to discern truth from fiction.

This apprehension is rooted in the observation that AI models can produce vast amounts of data that often resemble low-quality or nonsensical material. Given the exponential growth of such content across all social media channels, it’s tempting to assume that disinformation will similarly increase, overwhelming audiences with a deluge of unverified information.

However, I believe the situation isn’t so straightforward. Consider this analogy: If you or I pick up our smartphones and spend time browsing TikTok—viewing 100 to 150 short clips—adding AI-generated videos doesn’t necessarily inflate the number of content pieces we consume. The volume remains relatively stable because our attention span and content intake are naturally limited. The same applies to the broader flow of information online.

Furthermore, it’s important to recognize that humans already generate an enormous amount of disinformation independently of AI. Whether it’s political propaganda, fake news, or manipulated videos, the scale of human-created falsehoods has been staggering for years. Adding another petabyte of AI-crafted misinformation doesn’t dramatically shift what we’re exposed to; it’s more of a continuation than a fundamental change.

From a consumption standpoint, the core distribution remains consistent. My typical viewing habits—whether watching cute cat videos, viral fails, or political commentary—are unlikely to expand or diminish significantly based on AI content. Our media preferences are somewhat fixed, and I doubt AI alters this behavior substantially.

That said, AI can introduce subtler forms of deception—such as doctored video clips or realistic-sounding but fabricated statements from celebrities or politicians. These manipulations can be more convincing and less obviously false than blatant lies. For example, a video showing a politician saying something provocative, but in reality, they never made that statement, can be more insidious because it’s harder to spot at first glance.

In conclusion, while AI has the potential to introduce sophisticated disinformation, I remain cautious about overstating its impact. The volume of content—disinformation included—has already been overwhelming, and human consumption patterns tend to be stable. Ultimately, AI-driven disinformation may add nuanced deception, but it