"Researchers analyzed 'in the wild' attacks on more than 2,000 production AI applications over the past three months, uncovering significant vulnerabilities in current GenAI algorithms."
"Of the successful attacks, 90% lead to sensitive data leaks, underscoring the need for stronger safeguards and real-time protection mechanisms against adversarial behaviors."
Collection
[
|
...
]