How AI-generated text is poisoning the internet

This year has been crazy for AI. If you’ve spent a lot of time online, you’ve probably come across images generated by AI systems like DALL-E 2 or Stable Diffusion, or jokes, essays, or other text written by ChatGPT, the latest embodiment of the OpenAI GPT-3 grand language model.

Sometimes it’s obvious that an image or piece of text was created by an AI. But increasingly, the output generated by these models can easily trick us into thinking it was made by a human. And great language patterns, in particular, are trusting bullshit: they create text that seems correct, but may actually be full of lies.

While it doesn’t matter if it’s just a little fun, it can have serious consequences if AI models are used to offer unfiltered health advice or provide other forms of information. important. AI systems could also make it stupidly easy to produce tons of misinformation, abuse and spam, distorting the information we consume and even our sense of reality. This could be particularly worrying around elections, for example.

The proliferation of these large, easily accessible language models raises an important question: how will we know if what we read online is written by a human or a machine? I just posted about the tools we currently have for spotting AI-generated text. Spoiler alert: today’s detection toolkit is woefully inadequate against ChatGPT.

But there is a more serious long-term implication. We may be witnessing, in real time, the birth of a snowball of bullshit.

Great language models are trained on datasets that are built by scraping the internet for text, including all the toxic, silly, fake, and malicious things humans have written online. The finished AI models regurgitate these falsehoods as facts, and their output is streamed everywhere online. Tech companies are scraping the internet again, scavenging AI-written text that they use to train bigger, more convincing models, which humans can use to generate even more nonsense before it’s scraped again and again. again, ad nauseam.

This problem – the AI ​​feeding on itself and producing increasingly cluttered output – extends to images. “The internet is now forever contaminated with images created by AI,” Mike Cook, an AI researcher at King’s College London, told my colleague Will Douglas Heaven in his new article on the future of AI models. generative.

“The images we made in 2022 will be part of any model that is made from now on.”