‘ChatGPT Tainted Memories’ Exploit Enables Command Injection in Atlas Browser
LayerX Security found a flaw in OpenAI’s ChatGPT Atlas browser that lets attackers inject commands into its memory, posing major security and phishing risks.
OpenAI
OpenAI Atlas Omnibox Is Vulnerable to Jailbreaks
Researchers have discovered that a prompt can be disguised as an url, and accepted by Atlas as an url in the omnibox.
The post OpenAI Atlas Omnibox Is Vulnerable to Jailbreaks appeared first on SecurityWeek.
ChatGPT Atlas: OpenAI launches web browser centered around its chatbot
Company’s AI-powered browser built around marquee bot is designed to provide more personalized web experienceOpenAI on Tuesday launched an AI-powered web browser built around its marquee chatbot.“Meet our new browser—ChatGPT Atlas,” a tweet from the co…
ChatGPT ‘upgrade’ giving more harmful answers than previously, tests find
Campaigners ‘deeply concerned’ about response to prompts about suicide, self-harm and eating disordersThe latest version of ChatGPT has produced more harmful answers to some prompts than an earlier iteration of the AI chatbot, in particular when asked …
OpenAI’s Guardrails Can Be Bypassed by Simple Prompt Injection Attack
Just weeks after its release, OpenAI’s Guardrails system was quickly bypassed by researchers. Read how simple prompt injection attacks fooled the system’s AI judges and exposed an ongoing security concern for OpenAI.
OpenAI Finds Growing Exploitation of AI Tools by Foreign Threat Groups
OpenAI’s new report warns hackers are combining multiple AI tools for cyberattacks, scams, and influence ops linked to China, Russia, and North Korea.
ShadowLeak Exploit Exposed Gmail Data Through ChatGPT Agent
Radware researchers revealed a service-side flaw in OpenAI’s ChatGPT. The ShadowLeak attack had used indirect prompt injection to bypass defences and leak sensitive data, but the issue has since been fixed.
Trump hosts US tech leaders at White House dinner – minus Elon Musk
Tesla CEO’s absence is marked departure from his constant presence at the White House in early days of Trump 2.0As Donald Trump hosted leaders from the biggest US tech companies at a lavish White House state dining room dinner on Thursday night, there …
Leaked ChatGPT Chats: Users Treat AI as Therapist, Lawyer, Confidant
Leaked ChatGPT chats reveal users sharing sensitive data, resumes, and seeking advice on mental health, exposing risks of…
First AI-Powered Ransomware PromptLock Targets Windows, Linux and macOS
ESET has identified PromptLock, the first AI-powered ransomware, using OpenAI models to generate scripts that target Windows, Linux…