Securing Generative AI: Understanding and Defending Against Prompt Injection
Details
Over the past few years, generative AI has moved incredibly fast. What started as experimental technology is now deeply integrated into products, workflows, and enterprise systems. Organizations are using AI to generate code, automate decisions, analyze data, and even interact directly with customers through chatbots and assistants.
But as with any powerful technology, the rapid adoption of generative AI also introduces a completely new set of security challenges. Unlike traditional applications, AI systems don’t just execute deterministic code they interact with data, prompts, models, APIs, and external systems. This creates new attack surfaces that traditional security controls were never designed to handle.
In this presentation, we’ll explore what generative AI security is, why it’s becoming critical for organizations, the major threats targeting AI systems today, and the practical steps we can take to secure AI applications. Through the most common form of attack Prompt Injection.
Key Takeaways
- Learn what AI Security is and why it is becoming increasingly important.
- Explore common risks in AI systems such as model manipulation, prompt injection.
- What are the potential consequences of prompt injection attack.
- How to prevent prompt injection: best practices, tips, and tricks.
📅 Date: 18 April 2026
🤝 Registration & Networking: 10:00 am – 10:50 am
⏰ Event Start: 11:00 am
🏁 Event End: 13:00 pm
🍽️ Lunch & Networking: 13:00 pm – 14:00 pm
📍 Location: AWS Skills Center, Cape Town
⚠️ Attendee Limit: 60
🎯 Who Should Attend: Developers, security engineers, DevSecOps practitioners, and anyone interested in understanding the security risks and considerations of AI systems.
Reserve your spot now
Join the Community!
If you haven't joined the Discord community, please do so! You can find us on Discord at: https://devseccon.io/discordcommunity
