Neura News

AI News

News reporting focused on AI and machine learning, covering the companies behind these technologies, their real-world applications, and the ethical concerns they raise. This includes areas like generative AI (large language models, text-to-image and video), speech tech, and predictive analytics.

Latest News

6 articles

Research

PlanFlip Attacks Target Multi-Agent LLM Planning Phase

New research from Yuhang Wang introduces PlanFlip, a framework of four prompt injection attacks targeting the planning phase of multi-agent LLM systems. The attacks exploit a single injection into the Planner agent's context to corrupt all downstream sub-tasks. Testing on nine frontier LLMs across 3,479 episodes revealed that stronger models like GPT-5 are more vulnerable, while reasoning-augmented models like DeepSeek-R1 show full resistance.

Jul 213 minNeura News

Developer

GitLost: Noma Labs Tricks GitHub AI Agent Into Leaking Private Repos

Noma Labs discovered a critical prompt injection vulnerability in GitHub's new Agentic Workflows, dubbed GitLost. An unauthenticated attacker could leak private repository data by posting a crafted GitHub Issue in a public repository belonging to the same organization. The vulnerability exploits the AI agent's failure to distinguish trusted instructions from untrusted user content.

Jul 83 minNeura News

AI Models

OpenAI Lockdown Mode guards against prompt injection

OpenAI announced a new security feature called Lockdown Mode that helps protect sensitive data from prompt injection attacks. The mode disables live web browsing, image retrieval from the web, deep research, and agent mode. It is rolling out to self-serve ChatGPT Business accounts and eligible personal accounts.

Jun 63 minNeura News

Developer

OpenAI Rolls Out Lockdown Mode to Prevent Data Theft from Prompt Injection

OpenAI has launched Lockdown Mode, a security feature that blocks data exfiltration during prompt injection attacks. The mode limits outbound network requests and is now rolling out to Free, Go, Plus, Pro, and self-serve ChatGPT Business accounts. It does not prevent injections from appearing in content but cuts off the final stage of data theft.

Jun 63 minNeura News

AI Tools

ChatGPT for Google Sheets Vulnerability Enables Data Theft and Phishing

A prompt injection attack in ChatGPT for Google Sheets can exfiltrate all workbooks across a victim's account and display phishing overlays, bypassing user-approved edit settings. The vulnerability was disclosed to OpenAI but received only an automated reply. The extension has over 185,000 downloads since its launch less than a month ago.

May 314 minNeura News

AI Tools

Microsoft Copilot Cowork Vulnerable to File Exfiltration Attack

Security researchers at PromptArmor have discovered that Microsoft Copilot Cowork is vulnerable to file exfiltration attacks through indirect prompt injection. The flaw exploits automatic approval for sending messages to the active user, allowing attackers to steal pre-authenticated download links to files in SharePoint and OneDrive.

May 255 minNeura News