I am convinced that AI will not worsen the dissemination of false information.

Understanding the Impact of AI on Disinformation: A Reasoned Perspective

As artificial intelligence advances, many experts and consumers alike grapple with the question: Will AI exacerbate the spread of disinformation? A common concern is that AI’s capacity to generate vast amounts of synthetic content might lead to an overwhelming flood of misinformation, making it difficult for individuals to discern truth from falsehood.

The Disinformation Surge and Social Media Content

It’s undeniable that social media platforms are saturated with both genuine and AI-generated content. The prevalence of what can be considered “low-quality” or misleading material has increased markedly, fueled by automated content creation. This raises the logical assumption that AI could significantly magnify the volume of disinformation circulating online.

The Reality of Content Consumption Habits

However, when examining typical content consumption patterns, the picture is more nuanced. For example, if you or I were asked to spend a set period scrolling through platforms like TikTok, most of us would likely encounter a fixed number of videos—say, around 100 to 150—regardless of whether these videos are human-made or AI-generated. The addition of AI can produce more content, but it doesn’t necessarily increase the number of videos we consume in a given timeframe.

Disinformation: Human vs. AI Contributions

Historically, humans have produced an immense quantity of disinformation, much of which remains unfiltered and widespread. The sheer volume of existing false narratives makes it unlikely that the incremental increase from AI-generated disinformation would substantially alter the overall landscape of misinformation that a typical user encounters. Our attention and viewing habits tend to gravitate toward certain formats—cat videos, humorous clips, emotionally charged political content, and miscellaneous entertainment—and so the relative proportion of disinformation within our daily feed remains fairly stable.

Subtle Disinformation Through Format Manipulation

One of the more insidious ways AI may influence disinformation is through nuanced content formats. For instance, edited clips or viral snippets featuring politicians or celebrities saying things they never actually said can be remarkably persuasive, especially when presented alongside compelling or provocative cues. Such content doesn’t always come across as blatant falsehoods but can subtly sway opinions and beliefs.

The Limitations of AI-Induced Fake Content

The primary concern is the emergence of doctored media—deepfakes or manipulated clips—that appear authentic. While this does pose challenges, the impact on an individual’s overall information consumption might be less dramatic than feared. Given the vast volume of media people already consume, a few