Cybersecurity startup Astra Security has found serious vulnerabilities in more than half of the artificial intelligence (AI) applications it tested, particularly on fintech and healthcare platforms. The findings were presented at CERT-In Samvaad 2025, a government-backed cybersecurity conference.
The research outlines how large language models (LLMs) can be manipulated through prompt injections, indirect prompt injections, jailbreaks, and other attack methods. These tricks can cause AI systems to leak sensitive data or make dangerous errors.
In one…
