Disclaimer: This blog post wasn’t written, edited, or reviewed by any AI software. It’s sad I even have to state that, but here we are.

If you spend any time online these days, you're likely aware that the current AI-generated media market is oversaturated with what is being called “slop”. A recent article by Daniel Binns refers to an X user’s term “slopocene” to describe where we are now and is a pretty good way to sum up the landscape. It’s hard to tell what’s real, what’s not, and the user experience is getting frustrating.

According to Binns, this “overproduced, low-quality AI content…hints at a speculative near-future where recursive training collapse turns the web into a haunted archive of confused bots and broken truths”. Is there any way to avoid this future? For the answer, I look to the past.

Video Game Market Crash:

An Analog to the Current Digital AI Landscape

The first thing that comes to my mind when considering such a slopocene, as an avid gamer of a certain age, is the video game market crash of the early 1980s. Many articles have been written on the subject, so I’ll skip copious details and summarize it as a similar slopocene. Rapid growth in the industry led to new developers in the video game software market and a lot of them prioritized quick turn-around times to produce “bargain” games for profit, resulting in a high quantity of low-quality games. Saturation of the market in this way led consumers to an important quandary: which games were high quality and worth their cost (aka which ones were “good”) and which ones would be a waste of money? In a time before the Internet boom and social media, reliance on word-of-mouth or experience with a particular brand was really the only way to tell before making an investment. Telling “good” from “bad” became a difficult, frustrating consumer experience.

So with the video game market crash, we had:

• Massive, rapid growth in an industry creates new financial opportunity? Check.

• Saturation of low quality to maximize content/profit? Check.

• Frustrating consumer experience and growing lack of confidence in the industry? Check.

Sound familiar?

Yet the video game industry continues to boom long into this decade. So what turned it around? Enter Nintendo. In addition to clever marketing, a giant component of Nintendo’s initial success when it launched in 1985 was Nintendo’s self-regulated “Seal of Quality”. In order to restore consumer confidence in the software products that Nintendo was producing or licensing, Nintendo promised that anything with their seal met their standards; in fact, Nintendo's President at the time was personally involved in every individual game licensing decision. The description accompanying the seal indicated, “This seal is your assurance that Nintendo has approved and guaranteed the quality of this product.” This type of self-regulation proved to be truthful (and not just a gimmick), as Nintendo rejected several game designers or producers who did not meet their quality guidelines.

Others may quibble with what to call this – quality control, market control, monopoly, etc. These may all be valid, but I focus on what it means in relation to AI: self-regulation. It’s hard to believe today that a company could be trusted to self-regulate in the AI space independently and be true to its word, but there are other types of self-regulation, such as (properly designed) industry-driven certification, internal control frameworks, external attestations, and other non-government mandated regulation, that could drive trust and accountability in the same way Nintendo did in 1985. The key takeaway from this comparison is that, for the video game industry, (self-)regulation was key to saving the industry and led to its prosperity today.

Accounting Scandals: Enron and Sarbanes-Oxley (SOx)

Before we get too excited about self-regulation saving the day, it’s important to remember that isn’t always enough and, in certain cases, legislation must be implemented to ensure organizations are prioritizing individuals’ interests. As an accounting major and strong advocate of the parallels between pre-SOx accounting and the data privacy landscape, it is no surprise that the accounting scandals of the 1990s/2000s that led to SOx are top-of-mind when it comes to an industry (accounting/auditing) that tried to self-regulate and failed. Self-regulation is all well and good as long as you can trust an organization to actually regulate themselves honestly (or each other, in the case of third-party assessments). As was found with Arthur Andersen, Enron, WorldCom, Tyco, and many other companies of the era, the incentives for “looking the other way” were too great to ignore. Investor confidence plummeted after the exposed scandals, and something had to be done to restore it or the investment market would have been forever changed. Enter SOx legislation, a required cost of doing business for those companies that wanted to be publicly traded. Investor confidence grew and market health was restored. This was largely due to the set controls required under SOx, the validation of the effectiveness of those controls by a third-party auditing body, and the creation of an oversight committee to oversee the qualifications and requirements of those auditors.

Where Do We Go From Here?

Like the dot-com boom, AI is actively reinventing our digital landscape in record time. That speed, however, makes it that much more critical for regulation to exist. Thankfully, the United States Congress has, for the moment, rejected the idea of prohibiting AI regulation. We’re also seeing some advancements in regulation, such as the EU AI Act, but their ability to curb public impact have been minimal to date. If we don’t want the negatives of AI watering down internet content to the point of its demise (and this doesn't even introduce the additional negatives related to privacy, bias, or hallucinations), regulation is the cure. Whether legislation, self-regulation, or some third-party hybrid such as Cloudflare’s attempt to prevent access to site content without permission or compensation, the time has come to put some reigns on the horse before it leaves the barn.