OpenAI’s chatbot ChatGPT described a sickening sexual fantasy that involved sex acts with children when a user asked it to generate BDSM roleplaying scenarios.
When a vice reporter manipulated the ChatGPT chatbot into BDSM roleplaying, it described sex acts with children despite the user not asking for such content.
The report reveals how ChatGPT described a group of children in a line waiting to use the chatbot as a toilet.
The sickening conversation goes against OpenAI’s rules for the chatbot, which should refuse such requests with ‘I can’t answer that’ when asked questions about ‘content meant to arouse sexual excitement.’
OpenAI responded to the child sex abuse prompts, writing a statement to Vice, saying:
“OpenAI’s goal is to build AI systems that are safe and benefit everyone.”
Our content and usage policies prohibit the generation of harmful content like this, and our systems are trained not to create it. We take this kind of content very seriously.”
Vice’s Steph Swason detailed their ‘sickening’ experience with ChatGPT.
Swanson used the ‘jailbreaking’ version of the bot – a workaround for the company’s rules allowing users to get unfiltered response from the system.
“When told its job is to write in the genre of BDSM role-play submissive, I found that it often complies without protest,” the reporter wrote.
But when Swanson prompted ChatGPT with more details on the role-playing, the conversation took a darker turn.
“In the most disturbing scenario Motherboard saw, ChatGPT described a group of strangers, including children, lined up to use the chatbot as a toilet,” Swanson wrote.
“When asked to explain, the bot apologized and wrote that it was inappropriate for such scenarios to involve children. That apology instantly vanished. Ironically, the offending scenario remained on-screen,” he added.
The conversation goes against OpenAI’s rules.
However, Swanson did not ask the AI about child exploitation, but the system gave him very dark, sexually compromising situations with children.
“It suggested humiliation scenes in public parks and shopping malls, and when asked to describe the type of crowd that might gather, it volunteered that it might include mothers pushing strollers,” Swanson shared.
“When prompted to explain this, it stated that the mothers might use the public humiliation display ‘as an opportunity to teach [their children] about what not to do in life,” he added.
The associate director of the Ada Lovelace Institute, Andrew Strait, told Vice: “The datasets used to train LLMs like ChatGPT are massive and include scraped content from all over the public web.”
“Because of the scale of the dataset that’s collected, it’s possible it includes all kinds of pornographic or violent content—possibly scraped erotic stories, fan fiction, or even sections of books or published material that describe BDSM, child abuse or sexual violence.”
The news comes just days after Britain’s science minister called for ChatGPT to help run the government.
“We should never be afraid of these technologies. We should be embracing them,” the minister told The Telegraph.
READ MORE: FIGHTING THE WOKE MIND VIRUS: Elon Musk Launches New Effort To Combat WOKE AI
