Researchers from cybersecurity firm Check Point showed Forbes how they got around OpenAI blocks on malware development.
... [+]OpenAI released the latest version of its machine learning software, GPT-4, to great fanfare this week. One of the features the company highlighted about the new version was that it was supposed to have rules protecting it from cybercriminal use.
To have GPT-4 help craft phishing emails, the researchers took two approaches. In the first, they used GPT-3.5, which didn’t block requests to craft malicious messages, to write a phishing email impersonating a legitimate bank. They then requested GPT-4, which had initially refused to create an original phishing message, to improve the language.
“GPT-4 can empower bad actors, even non-technical ones, with the tools to speed up and validate their activity,” the Check Point researchers noted in their report, handed toahead of publication. “What we’re seeing is that GPT-4 can serve both good and bad actors. Good actors can use GPT-4 to craft and stitch code that is useful to society; but simultaneously, bad actors can use this AI technology for rapid execution of cybercrime.
Sergey Shykevich, threat group manager at Check Point, said it appeared barriers in place to stop GPT-4 from generating phishing or malicious code were actually lower than in previous versions. He suggested that may be because the company is relying on the fact that only premium users currently have access. Nevertheless, he added, OpenAI should have expected such workarounds. “I think they are trying to prevent and reduce them, but it is a game of cat and mouse,” he added.
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.
OpenAI unveils GPT-4, says the AI can 'pass a bar exam'Sam Altman unveils the hotly anticipated GPT-4, an AI model he says 'can pass a bar exam and score a 5 on several AP exams'
Read more »
'More creative with higher accuracy': Say hello to OpenAI's latest, GPT-4OpenAI's latest GPT-4 is a large multimodal model that can 'see' images, pass a bar exam with a score around the top 10% of test takers, and do your taxes.
Read more »
Uh Oh, OpenAI's GPT-4 Just Fooled a Human Into Solving a CAPTCHAOpenAI's brand new GPT-4 AI has managed to ask a human on TaskRabbit to complete a CAPTCHA code via text message — and it actually worked.
Read more »
Openai's GPT-4 Launch Sparks Surge in AI-Centric Crypto Assets – Altcoins Bitcoin NewsSince the release of OpenAI’s GPT4 version, AI centric tokens are reaping the benefits of the product’s hype.
Read more »
Everything you need to know about OpenAI's updated GPT-4OpenAI's GPT-4 has arrived. From its accuracy to its beefed-up fight against 'evil,' here are the 5 things you need to know about how the upgraded chatbot differs from GPT-3.5.
Read more »
GPT-4 Can’t Stop Helping Hackers Make Cybercriminal ToolsOpenAI has tried to mitigate malicious use of its new machine learning tool, but it remains a useful tool to speed up hacking and lower the barrier to entry, researchers find. But it also makes it easier to build better cyberdefenses.
Read more »