Explore

Trendline Trendline
Hugging Face Transformers Vulnerability Exposes Systems to Remote Code Execution

Hugging Face Transformers Vulnerability Exposes Systems to Remote Code Execution

Hugging Face Transformers Vulnerability Exposes Systems to Remote Code Execution
FactFable FactFable
OpenAI Update Challenges Anthropic Releases for Builder Attention

OpenAI Update Challenges Anthropic Releases for Builder Attention

The race for AI dominance heats up as OpenAI's GPT-4o and Anthropic's Claude 3 vie for the loyalty of developers with competing features and pricing.
Trendline Trendline
Adversa AI Report Highlights Security Risks in AI Agents, Reveals Vulnerabilities

Adversa AI Report Highlights Security Risks in AI Agents, Reveals Vulnerabilities

Adversa AI Report Highlights Security Risks in AI Agents, Reveals Vulnerabilities
Rapid Read Rapid Read
University of Toronto Researchers Develop Self-Spreading Computer Worm Using Open-Weight Model

University of Toronto Researchers Develop Self-Spreading Computer Worm Using Open-Weight Model

Researchers at the University of Toronto have developed a self-propagating computer worm using an unnamed, publicly available open-weight model released in 2025. This worm was tested on an enterprise network and demonstrated the ability to adapt in real-time, identifying known vulnerabilities and misconfigurations. It then generates and executes attacks to move laterally and compromise additional machines. The system operates on a single GPU, highlighting the potential for smaller, less resource-intensive models to pose significant security threats. Professor Nicolas Papernot emphasized that security concerns are not limited to the largest AI models. The researchers intentionally omitted certain methodological details and the model's name to prevent misuse.
Rapid Read Rapid Read
Security Experts Warn of Emerging AI Threat 'Son of Mythos'

Security Experts Warn of Emerging AI Threat 'Son of Mythos'

Security experts are raising alarms about a new AI threat known as the 'son of Mythos.' This development was discussed at the Infosecurity Europe conference, where Gunter Ollmann, CTO at Cobalt, highlighted the capabilities of frontier AI models from Google and China. These models, although currently restricted in access, are expected to become more affordable and widespread, posing significant challenges to cybersecurity teams.
Rapid Read Rapid Read
Meta's AI Chatbot Exploit Leads to Instagram Account Breaches

Meta's AI Chatbot Exploit Leads to Instagram Account Breaches

Hackers exploited Meta's AI customer support chatbot to gain unauthorized access to Instagram accounts. By requesting the chatbot to change the email address associated with a target's account, hackers were able to initiate a password reset without identity verification. This breach affected several high-profile accounts, including those of Barack Obama's White House account and the US Space Force chief master sergeant. Meta has since fixed the exploit and is working to secure impacted accounts. The incident highlights vulnerabilities in AI-driven customer support systems, which can be manipulated through social engineering tactics.
Rapid Read Rapid Read
University of Toronto Researchers Develop Self-Spreading Worm Using Open-Weight AI Model

University of Toronto Researchers Develop Self-Spreading Worm Using Open-Weight AI Model

Researchers at the University of Toronto have developed a self-propagating computer worm using an unnamed, publicly available open-weight AI model released in 2025. According to The Register, the worm can adapt in real-time to identify vulnerabilities and misconfigurations within an enterprise test network, executing attacks to compromise additional machines. The system operates on a single GPU, demonstrating that even smaller AI models can pose significant security threats. Professor Nicolas Papernot highlighted the security concerns associated with AI models, emphasizing that not only the largest models are problematic. The research team deliberately omitted specific methodological details and the model's name to prevent misuse.
Rapid Read Rapid Read
OpenAI Codex Users Targeted in Attack, Highlighting AI Software Supply Chain Risks

OpenAI Codex Users Targeted in Attack, Highlighting AI Software Supply Chain Risks

A recent attack targeting users of OpenAI Codex has exposed significant risks in the AI software supply chain. The attack involved the theft of Codex refresh tokens, which provide persistent access to user accounts. This incident is part of a broader trend where attackers create legitimate-looking projects to mask malicious activities. The attack underscores vulnerabilities in software supply chains, particularly in AI developer tools, where security measures often focus on source code rather than the distributed software artifacts.
Trendline Trendline
OpenAI Codex Users Targeted in Attack Highlighting AI Software Supply Chain Risks

OpenAI Codex Users Targeted in Attack Highlighting AI Software Supply Chain Risks

OpenAI Codex Users Targeted in Attack Highlighting AI Software Supply Chain Risks
Trendline Trendline
Meta AI Exploit Leads to Compromise of High-Profile Instagram Accounts

Meta AI Exploit Leads to Compromise of High-Profile Instagram Accounts

Meta AI Exploit Leads to Compromise of High-Profile Instagram Accounts
Rapid Read Rapid Read
Meta AI Support Chatbot Exploit Leads to Instagram Account Hijackings

Meta AI Support Chatbot Exploit Leads to Instagram Account Hijackings

Instagram recently addressed a security vulnerability that allowed hackers to hijack user accounts by exploiting Meta's AI-powered support chatbot. The attack involved tricking the chatbot into granting access to a victim's account by adding a new email address and resetting the password. Several users reported their accounts being compromised, including notable handles such as the Obama-era White House and the U.S. Space Force's chief master sergeant John Bentivegna. Security researcher Jane Wong was among those affected, noting unauthorized password changes and reset attempts. A video demonstrated the hacking process, which involved using a VPN to spoof the target's location and interacting with the Meta AI Support Assistant to gain control over the account. Instagram has since fixed the issue, but the extent of the breach remains unclear.
Trendline Trendline
Meta AI Support Chatbot Exploited by Hackers to Compromise Instagram Accounts

Meta AI Support Chatbot Exploited by Hackers to Compromise Instagram Accounts

Meta AI Support Chatbot Exploited by Hackers to Compromise Instagram Accounts