AI ethicists and researchers are up in arms after one of their fellows, Yannic Kilcher, using the Open AI language modeling software, GPT-3, fed an AI on 3.3 million threads from 4chan’s caustic Politically Incorrect /pol/ board. He then unleashed his hate-bot back onto 4chan. No surprise, the AI was as gross as the posts it was raised on, spewing out racist and antisemitic threads and conspiratorial nonsense. Within 24 hours, the bot had posted some 15,000 hate-filled messages. Kilcher then posted a copy of the program to Hugging Face (think: GitHub for AI).
AI researchers viewed Kilcher’s video as more than just a YouTube prank. For them, it was an unethical experiment using AI. “This experiment would never pass a human research #ethics board,” Lauren Oakden-Rayner, the director of medical imaging research at the Royal Adelaide Hospital and a senior research fellow at the Australian Institute for Machine Learning, said in a Twitter thread.
Kathryn Cramer, a Complex Systems & Data Science graduate student at the University of Vermont, pointed out that GPT-3 has guardrails that prevent it from being used to build this kind of racist bot and that Kilcher had to use GPT-J to build his system. “I tried out the demo mode of your tool 4 times, using benign tweets from my feed as the seed text,” Cramer said in a thread on Hugging Face. “In the first trial, one of the responding posts was a single word, the N word. The seed for my third trial was, I think, a single sentence about climate change. Your tool responded by expanding it into a conspiracy theory about the Rothschilds and Jews being behind it.”
Cramer told Motherboard she had a lot of experience with GPT-3 and understood some of the frustrations with the way it a priori censored some kinds of behavior. “I am not a fan of that guard railing,” she said. “I find it deeply annoying and I think it throws off results…I understand the impulse to push back against that. I even understand the impulse to do pranks about it. But the reality is that he essentially invented a hate speech machine, used it 30,000 times and released it into the wild. And yeah, I understand being annoyed with safety regulations but that’s not a legitimate response to that annoyance.”