Our AI Dystopia: Part 2 of 2

Last week, I discussed how, as artificial intelligence becomes more capable and accessible, AI-generated media is fueling the spread of disinformation and misinformation. This is especially problematic in the United States this year as many of us will be seeking reliable information to inform our votes for congressional elections. AI media has made it much harder to find and verify the truth online.

But there have always been ways to combat the spread of falsehoods. If AI is making these falsehoods spread faster and more efficiently, then we need to scale up our solutions to combating lies as well. The best ways we can combat AI-generated media effectively are to increase our resistance to false information and to hold the entities that benefit from AI more accountable.

Media Literacy

In my blog last week, I left off by imploring you to take what you see and hear with a grain of salt. This attitude of skepticism is, in essence, the most accessible solution to AI-generated slop: media literacy.

Media literacy refers to your ability to critically analyze pieces of media, to understand their effect on your thoughts, feelings, and behavior, and to determine their accuracy and credibility. A media literate person asks “Does this media come from a credible source?”, “Can I identify bias in this media’s message?”, and “Are other, trustworthy sources agreeing with this media?”

While you may not be able to answer these questions all the time, simply asking them fosters a mindset ready to recognize false information more frequently, and to be slower to accept falsehoods that are less obvious. In the context of AI-generated media, having media literacy is that metaphorical “grain of salt” that allows you not only to accept the evidence of your eyes and ears, but to verify it.

Our first priority should be fostering our own media literacy. This takes time, since a large part of becoming more media literate is learning to recognize telltale signs of, and developing good habits that keep you alert to, bad information. A good place to start your training is to make a list of trustworthy sources of information that you can use to verify the factuality of a piece of media that catches your attention, whether that be an image or article. Tools like ad fontes’ media bias chart can help us determine what information sources are credible enough to be trusted.

But media literacy becomes far more effective when we encourage our peers to build it alongside us. Disinformation and misinformation are so ubiquitous at this point that it’s inevitable you’ll be fooled by something, but if we surround ourselves with media literate people, they can serve as a safety net to correct us when we’re wrong. So the next time you’re sent an AI-generated image by a well-meaning friend or relative, use it as an opportunity to help them become more media literate: explaining how you determined that image was fake instead of just telling them that it is. One day they may need to return the favor.

Legislation and Regulation

Training ourselves and our peers to become more media literate is an effective and accessible way to combat all forms of false information, but that doesn’t do much to protect our society as a whole from its dangers. For long term solutions to disinformation and misinformation, especially forms exacerbated by AI-generated media, we should push for legislation and regulation that combats it. For example, California has implemented a law dictating that learning media literacy be a part of K-12 education. By training youth to determine truth from falsehood, our society as a whole will become more resistant to all kinds of false information, including forms perpetuated by AI-media.

But the responsibility to control AI-generated media doesn’t rest solely on the common peoples’ shoulders. Many powerful entities like tech companies and governments benefit from AI technology’s progress and use, so it’s important to hold them accountable with proper legislation and regulation to prevent the technology from being misused at a large scale.

Determining exactly how AI regulation and legislation are implemented and what they should accomplish is challenging for a few key reasons:

AI is a rapidly evolving technology, so it can quickly outgrow rules meant to govern it
Since AI is a global technology, there is significant jurisdictional complexity to its regulation
There are currently no universally accepted standards for evaluating and certifying what “AI technology” is, so it’s difficult to make laws governing it

Despite the difficulties, however, it’s important to continue pushing for more governance over AI since it has a constantly growing impact on our lives; we can’t even perform a simple Google search anymore without seeing an AI overview.

Without proper regulations or legislation to answer to, tech companies will have little incentive to do anything that isn’t profitable. Initiatives like removing bias and misinformation from their products, making AI technology more transparent to the public, or stopping the use of sensitive consumer data to train their products without user permission (an entirely different problem that could be its own blog) won’t be seriously considered unless they make the company money.

Additionally, some of these tech companies run social media platforms, such as Meta and X. Social media is where the vast majority of AI-generated slop goes viral, and tech companies have similarly little incentive to control it so long as it promotes engagement that makes them money. It’s widely agreed that social media platforms could do a lot more to combat dis- and misinformation propagating on their platforms, such as employing independent, third-party fact checking and having more robust trusts and safety mechanisms to address false information.

In the United States, there is currently no federal legislation targeted towards regulating AI. Instead a far less effective patchwork of state-level legislation provides some idea for what proper regulation could look like, such as Colorado passing a law that “requires AI developers and deployers of high-risk AI systems to exercise reasonable care to prevent algorithmic discrimination” (Mind Foundry, 2026). Unfortunately, the current presidential administration seems reluctant to pass any sort of law governing AI for fear that it will stifle technological progress and inhibit the US’s ability to maintain competitiveness with other countries. They even reversed Biden’s AI Oversight Approach, an executive order meant to “impose reporting and safety obligations on AI companies, requiring them to test and disclose their models’ capabilities and risks.”

Ultimately, if we want AI to be governed by better legislation and regulations, we need to make governments and tech companies alike hear our demands. Perhaps you can start by writing to your senator urging them to prioritize AI-focused legislation or finding ways to support organizations lobbying for AI regulations. Federal-level regulation of AI is a much loftier goal than simple media literacy, but it’s one we can reach collectively.

Conclusion

At this point, we couldn’t stop AI’s growth and integration into society even if we wanted to. So we are left with two options: let it constantly make a fool of us or grow alongside it. I prefer the latter option. After all, AI was created to be a tool for humanity, we just haven’t learned to wield it quite yet.

If you think your company could use some help dealing with AI-related issues, we at Crimson Vista offer several tools that can help you, including VistaTier-X and SecurityHub. Let’s move into the age of AI together.

Our AI Dystopia: Part 2 of 2

Media Literacy

Legislation and Regulation

Conclusion

Sources

Keep reading.

Psychology-Aware Security Design

AI: The New Face of Cybersecurity

The Limitations of Hashing for Data Anonymization