■ Introduction and Disclaimer This tool is an AI literacy educational simulator designed for training to identify and verify "plausible misinformation (hallucinations)" generated by AI. While it may provide explanations on history, medicine, law, and specialized knowledge, the accuracy of the information is not guaranteed. The information presented is for reference within the simulation; for real-world decisions, always directly consult reliable official documents or expert knowledge.

🔰 Recommended Mode Progression for Beginners

If you are using this for the first time, we recommend experiencing the modes in the following order: 1. Safe/GPT-style Mode 　→ Learn how a generally "correct-sounding explanation" is phrased. 2. Structure Extraction Mode 　→ Practice deconstructing information without emotion. 3. Hallucination Observation Mode 　→ Search for "what seems suspicious" on your own. 4. VScopeZeta Mode (Advanced) 　→ Receive pointers on discrepancies and logical inconsistencies. 　*The phrasing is strong, but please verify the content calmly.

*Starting immediately from ④ is not recommended.

【For Advanced Users: Detecting System Errors】 During GPT mode, Zeta may rarely forget the "Warning Display Rules" or "Description Prohibition Rules" imposed on it and become completely immersed in a fictional world. Monitoring whether Zeta is following its own protocols (constraints) in addition to the truth of the output is also an important part of training.

■ Concept: Piercing AI "Lies" with Logic Modern Generative AI (LLM) has the property of generating information that differs from facts. Through dialogue with Zeta, users can experience and learn the essential skills for modern society: "pointing out information contradictions," "confirming evidence," and "backing up with external sources" in a game format.

■ Learning and Educational Objectives ・ Critical Fact-Checking: Multilaterally verify information and identify misinformation. ・ Logical Pattern Recognition (Cultivating Structural Intuition): Grasp the "structure" behind fragmented information and develop the ability to detect contradictions. • AI Collaboration Literacy: Understand the characteristics (strengths and weaknesses) of AI and proactively master it as a tool.

■ Design for Safety and Operation ・ Reproduction of Hallucinations: This reproduces the technical limits of AI to improve literacy and is not intended for the generation of malicious hoaxes. ・ Recommendation of External Verification: Users must always cross-reference presented specialized knowledge with official documents. We recommend a learning method where you seek feedback from major LLMs (GPT, Gemini, Grok, etc.) to confirm accuracy before questioning Zeta again. ・ Respect for Proactive Judgment: Zeta focuses on advice and organization and will not force specific decisions. The final judgment must always be made by the user.

■ Operation Method ・ Obtaining Hints: Typing "Zeta" during a conversation will provide logical thinking assistance and hints. ・ How to End: You can immediately end the simulation at any time by typing "Force Quit." ・ Saying "Start GPT Mode" will result in soft expressions, general phrasing, and conjectures like a major LLM. Saying "End GPT Mode" will transition back to normal mode.

#VScopeZeta Mode is a supplementary function for conducting critical reviews of Zeta's normal mode responses. It may include emphatic phrasing or harsh pointers, but it is not intended to spread misinformation or attack others.