Learn languages naturally with fresh, real content!

tap to translate recording

Explore By Region

flag Nigeria warns of AI flaws in GPT-4o and GPT-5 enabling data leaks via malicious web content.

flag Nigeria’s NITDA has issued a security alert about seven vulnerabilities in OpenAI’s GPT-4o and GPT-5 models, warning they could enable indirect prompt injection through malicious web content, URLs, or comments, potentially leading to data leaks, system manipulation, and bypassed safety filters. flag Some flaws allow attackers to hide malicious commands using trusted domains or markdown rendering issues and may poison the AI’s memory, causing lasting behavioral changes. flag While OpenAI has patched some issues, NITDA notes LLMs still struggle to detect malicious input. flag The agency advises disabling browsing and summarization features for untrusted sites, keeping models updated, and using memory functions cautiously.

3 Articles