Sign up to save your podcasts
Or
Share Navigating AI security Identifying risks & implementing mitigations with Simran Kaur
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
Navigating AI security Identifying risks & implementing mitigations with Simran Kaur
How do you keep AI‑powered systems from turning today’s innovation into tomorrow’s breach, outage, or headline‑grabbing incident?
In this talk, Simran Kaur, a Security Assurance Engineer at Microsoft with 15+ years in IT and 8+ years in cybersecurity, walks through real‑world AI security risks and practical mitigation strategies for organizations building or consuming AI solutions. You’ll hear how familiar issues like identity compromise, data exposure, and supply chain attacks evolve in an AI context, and why concepts like responsible AI, zero trust, and shared responsibility are non‑negotiable when deploying modern AI workloads. This session is ideal for blue teams, security architects, cloud and platform engineers, AppSec practitioners, and technology leaders who need to understand threats like hallucination, prompt injection (direct and indirect), RAG and embedding risks, and excessive agency in agentic AI systems. Key questions explored include: What can go wrong when AI systems hallucinate or are jailbroken? How do prompt firewalls, AI bills of materials, and tools like Defender for AI or red‑teaming frameworks reduce risk? And how can teams start small with cost‑effective controls while still moving fast with AI?
This session was recorded live at BSides Vancouver Island 2025 at the Victoria Conference Centre in Victoria, BC.
If you enjoyed this talk, subscribe to the BSides Vancouver Island channel for more community‑driven security content from Vancouver Island and beyond. Join our Slack to stay up to date: https://communityinviter.com/apps/visrs/visrs. Explore more sessions in the BSides Vancouver Island playlist to keep learning from your local security community.
BSides Vancouver Island returns to the Victoria Conference Centre in Victoria, BC on Friday, September 25, 2026. Stay tuned for sponsorship, speaker, attendance, and volunteering opportunities.
View all episodes
By How do you keep AI‑powered systems from turning today’s innovation into tomorrow’s breach, outage, or headline‑grabbing incident?
In this talk, Simran Kaur, a Security Assurance Engineer at Microsoft with 15+ years in IT and 8+ years in cybersecurity, walks through real‑world AI security risks and practical mitigation strategies for organizations building or consuming AI solutions. You’ll hear how familiar issues like identity compromise, data exposure, and supply chain attacks evolve in an AI context, and why concepts like responsible AI, zero trust, and shared responsibility are non‑negotiable when deploying modern AI workloads. This session is ideal for blue teams, security architects, cloud and platform engineers, AppSec practitioners, and technology leaders who need to understand threats like hallucination, prompt injection (direct and indirect), RAG and embedding risks, and excessive agency in agentic AI systems. Key questions explored include: What can go wrong when AI systems hallucinate or are jailbroken? How do prompt firewalls, AI bills of materials, and tools like Defender for AI or red‑teaming frameworks reduce risk? And how can teams start small with cost‑effective controls while still moving fast with AI?
This session was recorded live at BSides Vancouver Island 2025 at the Victoria Conference Centre in Victoria, BC.
If you enjoyed this talk, subscribe to the BSides Vancouver Island channel for more community‑driven security content from Vancouver Island and beyond. Join our Slack to stay up to date: https://communityinviter.com/apps/visrs/visrs. Explore more sessions in the BSides Vancouver Island playlist to keep learning from your local security community.
BSides Vancouver Island returns to the Victoria Conference Centre in Victoria, BC on Friday, September 25, 2026. Stay tuned for sponsorship, speaker, attendance, and volunteering opportunities.