Lesson 6 of 6 · 5 min read
Denial of Wallet: Cost as an Attack Surface
LLM applications pay per token, so attackers can harm them by driving cost rather than taking them offline: long prompts, maximum outputs, inflated reasoning, routing to the most expensive model, agent loops and stolen keys used to resell model access. OWASP's 2026 list moved unbounded consumption to sixth and asks for hard caps that stop inference, not just alerts. Effective controls budget money per key, user, tenant and team, estimate and limit tokens before a call, cap reasoning and output, watch the share of expensive models, and plan a break-glass process that does not itself become the attack path.
free sign-in, no card required
Practise it as an interviewer asks it
Walk me through red-teaming a customer-facing chatbot from scoping to report.What does the shared responsibility model mean for a managed AI service?How do you respond to suspected cross-tenant data leakage through an LLM feature?A startup shipped a chatbot with no governance. What do you set up first?What is unbounded consumption, or denial of wallet, and how do you prevent it?How can a chat interface become vulnerable to stored XSS through model output?