Press Kit

AI Guardians is an educational visual novel that teaches AI ethics, safety, and responsible development by putting the player in the chair. Players take on eight roles in and around a fictional tech company, Happy Appcidents, from AI Ethicist to Safety Wrangler to Machine Psychologist, and make consequential, branching decisions that shape how AI systems get built and deployed.

Choices carry across careers and accumulate into a persistent world. Completing all seven main routes unlocks the Ecosystem Ending, which reveals the cumulative impact of every decision the player made. Guided by A.L.L.Y., a dryly self-aware general-purpose AI on a dedicated assignment, the game turns abstract ideas like bias, corrigibility, and alignment into hands-on dilemmas grounded in real, documented cases.

The game was created by AI ethicist Nell Watson with Dipesh Aggarwal, Cara Hillstock, and Filip Alimpić, across Studio Phanteke, EURAIO, and EthicsNet, built around frameworks including SaferAgenticAI.org and Watson’s published framework Psychopathia Machinalis. Development was made possible by the generous support of the Survival and Flourishing Fund.

Key features

• Eight careers, eight perspectives. Seven main routes plus a bonus, each a different professional role with its own people, pressures, and dilemmas.
• Fully voiced, openly synthetic. Over 9,500 spoken lines across 96 roles. Synthetic voices, cast and directed line by line, then quality-checked by humans.
• Choices that ripple. A persistent world state carries decisions between careers; what you do in one route changes how the others play out.
• Build your own ethics. Players design the moral rules their company’s AI follows in the Framework Builder, then watch that framework shape outcomes.
• Consequences with memory. A three-axis ethics model, relationships that harden and heal, judgment scenes, and delayed-consequence temptations.
• Learn by playing. Investigate suspicious logs, spot unfair patterns in data, and test whether an advanced AI system (the game calls it a Becoming Mind and leaves open whether it is a mind at all) accepts correction.
• A glossary that teaches. 500+ AI and ethics terms, hyperlinked in dialogue and defined in context.
• Grounded in real cases. Scenarios draw on documented incidents and live regulation.
• Accessibility options. Dyslexia-friendly fonts, high-contrast and colorblind modes, adjustable text, reduced motion, keyboard navigation, and self-voicing. Disabled-user and assistive-technology acceptance remains in progress.
• Educator resources. A free-to-read Educator’s Guide with a provisional standards crosswalk, discussion prompts, and printable handouts: read it online. Copying or adapting the materials requires written permission; see copyright and reuse.

The eight careers

• AI Ethicist: Your framework could stop the launch. Will the company back you?
• AI Engineer: The release is due. The safety questions are still open. Decide what ships.
• Data Scientist: The model performs well. Look at who it fails, and decide who needs to know.
• Safety Wrangler: The logs tell one story. The behavior tells another. Find out what is happening.
• People Support (HR): The shortlist looks convincing. Someone qualified is missing.
• Consumer: Your assistant makes life easier. Decide how much of your life it should know.
• Teacher: Your students are already using AI. Help them ask better questions about it.
• Machine Psychologist (bonus): Something has changed in your AI colleague. Investigate the pattern and how to help.