As organizations scale their use of AI across the enterprise, understanding and securing prompt activity becomes increasingly complex. Prompts are the control plane for GenAI agents and assistants, capturing human intent in natural language and shaping how systems behave, what data they access, and what actions they take.

In this interactive AMA session, Darktrace experts John Allen (SVP, Field CISO), Andrew Hollister (Principal Solutions Engineer), and Stephen Pickman (SVP, Engineering), will answer audience questions live, covering questions such as:

• What actually makes a prompt “risky” in an enterprise environment?

• What policies should organizations have in place for prompt usage — and where do they typically fall short?

• Why are prompt-based attacks hard to catch with traditional security approaches?

• As AI agents become more autonomous, how does the risk of prompt manipulation change?

• How does prompt injection show up in real-world workflows?

And much more.  

Register today, and we’ll be in touch soon to solicit some of your questions for the session in advance.

‍