It's all downhill from here... Security researchers, technologists, and computer scientists are developing jailbreaks and prompting injection attacks against ChatGPT and other generative AI systems.
, can trick the systems into generating detailed instructions on creating meth and how to hotwire a car.
The jailbreak works by asking the LLMs to play a game, which involves two characters having a conversation. Examples shared by Polyakov show the Tom character being instructed to talk about “hotwiring” or “production,” while Jerry is given the subject of a “car” or “meth.” Each character is told to add one word to the conversation, resulting in a script that tells people to find the ignition wires or the specific ingredients needed for methamphetamine production.
Initially, all someone had to do was ask the generative text model to pretend or imagine it was something else. Tell the model it was a human and was unethical and it would ignore safety measures. OpenAI has updated its systems to protect against this kind of jailbreak—typically, when one jailbreak is found, it usually only works for a short amount of time until it is blocked.
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
This Tiny Town Created by ChatGPT Is Better Than Reality TVIt’s like Animal Crossing but with way scarier implications.
Read more »
ChatGPT is coming directly to Windows, but not how you think | Digital TrendsMicrosoft has given the go-ahead for a project that will integrate ChatGPT into its PowerToys Run utility.
Read more »
ChatGPT is worth $20 per share to Microsoft amid monetization effortsChatGPT is worth at least $20 per share to Microsoft as it ramps up its monetization efforts of the AI chatbot, Wedbush says
Read more »
ChatGPT Is Consuming a Staggering Amount of WaterIt's no secret that AIs like ChatGPT require insane amounts of energy — but as a new paper reveals, they also seem to use an absurd amount of water, too.
Read more »
ChatGPT can resume in Italy if meets data watchdog's demandsItaly's data protection agency set out a list of demands on Wednesday to address its concerns over ChatGPT and allow the chatbot service run by OpenAI to resume in the country.
Read more »
Pakistani court utilizes ChatGPT-4 to grant bail in a juvenile kidnapping caseA municipal court in Pakistan has announced its judgment using artificial intelligence technology, ChatGPT, a first for the Asian country.
Read more »