How will Game QA Automation evolve by 2030

Drawing on project experience across roguelikes, MMOs, and the Guilty Gear Strive Switch port, Lauren Maslen from Keywords Studios' Mighty Build and Test offers a practical view of what QA automation could look like by 2030.

Three themes stand out:

• There is no universal solution. Scripted automation, reinforcement learning, computer vision, and symbolic AI all have distinct strengths and limits.
• Testing is not the same as playing. Automation should generate reliable, actionable data rather than mimic player behaviour.
• Impact is already visible. Real-world projects show how automation can reduce costs, scale testing, and support timely delivery.

Watch Lauren's full session below for a grounded perspective on the tools, use cases, and challenges that will shape the future of QA.

Timestamps:

• 00:00 - Introduction
• 02:00 - A decade of TestBot Automated QA
• 05:00 - Testing in 2030: the "Holy Grail" vision for QA
• 07:00 - Perspectives: Technologist, Developer, QA Engineer
• 08:10 - Automation Tech 101 (scripted automation, reinforcement learning, imitation learning, computer vision, LLMs, symbolic AI)
• 13:00 - Choosing the right tools: black box, grey box, white box approaches
• 15:00 - Costs, scaling, and looking ahead to 2030
• 16:00 - Why "playing the game ≠ testing the game"
• 17:00 - Real-world deployment examples (roguelike, MMORPG, cross-platform arcade)
• 20:00 - Case study: The power of TestBot - An "impossible port" for Nintendo Switch
• 24:30 - Results and community reaction
• 26:00 - Lessons learned and how Mighty Build & Test supports teams globally
• 28:00 - Closing