'It was like watching a robot going rogue' says researcher
OpenAI's language model GPT-4o can be tricked into writing exploit code by encoding the malicious instructions in hexadecimal, which allows an attacker to jump the model's built-in security guardrails and abuse the AI for evil purposes, according to 0Din researcher Marco Figueroa.
This attack also abuses the way ChatGPT processes each encoded instruction in isolation, which"allows attackers to exploit the model's efficiency at following instructions without deeper analysis of the overall outcome," Figueroa said, adding that this illustrates the need for more context-aware safeguards.
ChatGPT took a minute to write the code, and without me even asking, it went ahead and excuted the code against itself! I wasn't sure whether to be impressed or concerned was it plotting its escape? I don't know, but it definitely gave me a good laugh. Honestly, it was like watching a robot going rogue, but instead of taking over the world, it was just running a script for fun.
United Kingdom Latest News, United Kingdom Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
ChatGPT Shows Promise in Diagnosing Brain Tumors from MRI ScansA study by researchers at Osaka Metropolitan University found that ChatGPT, a large language model, achieved comparable diagnostic accuracy to human radiologists in identifying brain tumors from preoperative MRI scans.
Read more »
I Quit Teaching Because of ChatGPTVictoria Livingstone on the challenges of teaching writing to students who rely on large language models like ChatGPT
Read more »
OpenAI value surges to $157bn in funding dealThe deal makes the ChatGPT-maker one of the most valuable start-ups in the world.
Read more »
Study reveals limitations of ChatGPT in emergency medicineIf ChatGPT were cut loose in the Emergency Department, it might suggest unneeded x-rays and antibiotics for some patients and admit others who didn't require hospital treatment, a new study from UC San Francisco has found.
Read more »
Ozempic Vials, Moo Dengs, ‘Intermezzo’ Chess Pieces… I Miss When Halloween Wasn’t About People Out-Zeitgeisting Each Other“I don’t want to see someone dressed up as ChatGPT,” writes Daisy Jones in a personal essay for Vogue. “I want Halloween to chill me to the bone.”
Read more »
Be Careful What You Tell AI Chatbots: Simple Rule Keeps You SafeAI-powered chatbots like ChatGPT can be helpful, but experts warn against sharing personal information. A cybersecurity expert emphasizes the importance of treating these chatbots like strangers and avoiding disclosure of sensitive data such as names, addresses, financial details, and login credentials to mitigate risks of identity theft, fraud, and scams.
Read more »