The biggest AI flops of 2024

AI penetration has penetrated into almost every corner of the Internet

generative artificial intelligence Makes it easy to create large amounts of text, images, videos, and other types of materials. Because the model you choose takes just seconds to output results after entering a prompt, these models have become a fast and easy way to generate content at scale. In 2024, we start calling this (often poorer quality) media: AI slop.

This low-risk way of creating AI spillovers means it can now be found in almost every corner of the internet, from Newsletter In your inbox and books for sale Amazon Advertisements and articles online and poor quality images on social media. The more emotions these photos evoke (wounded veterans, crying childone support signal in the Israeli-Palestinian conflict), the more likely they are to be shared, resulting in higher engagement and ad revenue for savvy creators.

Not only is the overflow of artificial intelligence troubling, its rise also raises real questions for the future of the models that helped produce it. Since these models are trained on data scraped from the web, the growing number of spam sites containing AI garbage means there is a very real danger to the models. Output and performance will gradually deteriorate.

Artificial intelligence art is distorting our expectations of real events

This was also the year when the impact of hyper-realistic artificial intelligence imagery began to permeate our real lives. Willy’s Chocolate Experiencean unofficial immersive event inspired by Roald Dahl charlie and the chocolate factorywhich made headlines around the world in February after its fantastic artificial intelligence-generated marketing materials impressed visitors that it was far more grandiose than the sparsely decorated warehouses created by its producers.

Likewise, hundreds of people lined the streets of Dublin to wait A non-existent Halloween parade. A Pakistani website used artificial intelligence to create a list of events in the city before October 31, which was shared widely on social media. All illustrate how misguided public trust in artificial intelligence-generated materials online may come back to haunt us.

Grok allows users to create images of almost any scene

Most major AI image generators have guardrails — rules that dictate what AI models can and cannot do — to prevent users from creating violent, explicit, illegal, and otherwise harmful content. Sometimes these guardrails are simply to ensure that no one is blatantly using someone else’s intellectual property rights. But Grok, the assistant built by Elon Musk’s artificial intelligence company xAI, ignores almost all of these principles, which is consistent with Musk’s rejection of what he calls “awakened artificial intelligence.”