Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore how modern AI risks represent amplified versions of familiar attack patterns in this 30-minute conference talk that reframes enterprise AI security challenges. Examine prompt injection vulnerabilities and slopsquatting in AI-generated code while discovering how trusted enterprise workflows face exploitation at unprecedented scale and speed. Analyze two detailed real-world threat scenarios: data exfiltration through compromised prompts and malicious dependency infiltration in AI-assisted development pipelines. Understand how blind trust in large language models, shadow AI implementations, and inadequate data governance create enterprise-wide security risks. Learn practical approaches using automated agents, data provenance tracking, and behavioral monitoring to detect and contain threats before they achieve dangerous scale. Gain insights into protecting AI-integrated workflows and maintaining security posture as artificial intelligence becomes increasingly embedded in enterprise operations.
