Let the Agents Run? The Case for (and Against) Full Automation
18 Jan 2027
WorldGaming Stage
Scaling a Stronger Technology Stack (Paid)
Generative AI helped gaming companies write, summarise and assist. The evolution - agentic AI - takes action: it resolves support tickets, flags and freezes suspicious accounts, scales infrastructure before major events, and runs compliance checks around the clock, amongst other functions. But not every process should run without a human. In this fireside chat, AWS shares what it’s seeing across gaming and other high-volume industries: where full automation creates real value, where human oversight still matters, and what architecture is needed to run AI agents reliably, securely and within regulatory limits at peak scale.
- The practical difference between generative and agentic AI, and why it matters for gaming operations
- A simple way to decide what to automate fully, what to keep human-assisted, and what to leave alone
- High-value use cases for agentic AI
- Guardrails for regulated markets: auditability, explainability, and keeping a human able to step in
- Scaling foundations: data readiness, cost control, latency, and resilience during traffic spikes