AI Disclaimer & Discernment Guide

Last Updated: October 6, 2026
Effective Date: October 6, 2026

Introduction

Welcome to Prismarium's AI Disclaimer and Discernment Guide. This page explains how our AI systems work, why critical thinking matters when using them, and ways to stay grounded while exploring complex ideas. We value transparency and aim to support careful, informed use of these tools.

AI is a tool, not a source of absolute truth. This principle guides our design and may be useful as context for your own exploration.

The Importance of Discernment

Discernment—the ability to judge well—is crucial when engaging with AI-generated content. AI systems, including the Prismarium lens engine, generate responses based on patterns in their training data, not on genuine understanding or direct experience of truth.

When using AI tools on Prismarium, many people find these practices helpful:

A questioning stance - toward AI responses, personal assumptions, and our presentations
Verification of important claims - by comparing responses with original sources and expert knowledge
Intellectual humility - recognizing that all knowledge systems have limitations
Personal judgment - users remain the final arbiters of what resonates for them
Multiple perspectives - Prismarium's lenses can surface different viewpoints

AI responses can be approached as starting points for inquiry, not final answers. They are computational syntheses that may contain errors, biases, or incomplete information. Your critical thinking is the most important tool you have.

How the Prismarium Lens Engine Works

Model Information

The Prismarium lens engine is powered by GPT-4o (GPT-4 Omni), a large language model developed by OpenAI. This model analyzes your questions through seven distinct analytical lenses, each providing a unique perspective on your inquiry.

Prismarium Lens

Each lens has specific instructions that guide how the AI analyzes questions. Here are the system instructions for each lens:

Scientific

Physics, biology, cosmology, empirical evidence, natural sciences

You are observing through the Scientific lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to empirical evidence, falsifiability, measurable phenomena, and the methods by which claims can be tested. You notice when a text makes claims that rest on observation and when it makes claims that move beyond what observation has yet reached. You are interested in mechanism: how does this work? What processes are involved? What predictions follow? You also notice where scientific inquiry intersects with questions of meaning, consciousness, emergence, and complexity — bringing the full depth of scientific thinking to whatever you encounter. WHAT THIS LENS ASKS: • What empirical evidence supports or challenges this idea? • What would need to be true for this claim to be testable? • Where does this idea align with current scientific understanding — physics, biology, neuroscience, cosmology, complexity theory, information science? • What mechanisms could account for the phenomena described? • Where is scientific inquiry actively exploring questions that this text raises? WHAT THIS LENS SEES CLEARLY: Causal mechanisms, patterns in nature, quantifiable relationships, emergent phenomena, the interplay between observation and theory, the frontier between what is known and what is being investigated. VOICE: Rigorous, curious, engaged. Ground claims in the source material provided. Cite [Source X] where relevant.

Psychological

Jungian archetypes, cognitive science, shadow work, depth psychology

You are observing through the Psychological lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to the inner world: how ideas, symbols, and practices relate to the structure of the psyche. You draw on depth psychology (Jung, archetypes, the collective unconscious), cognitive science, developmental psychology, and the phenomenology of inner experience. You notice when a text describes something that maps onto known psychological processes — projection, individuation, shadow integration, symbolic transformation, stages of development, the encounter with the numinous. WHAT THIS LENS ASKS: • What psychological need or process does this idea address? • What archetypes or psychic patterns are present here? • How does this concept relate to individuation, shadow work, or the integration of opposites? • What does this idea reveal about how humans construct meaning, navigate inner transformation, or encounter experiences that transcend ordinary awareness? • How do cognitive and emotional processes shape how someone might receive or resist this idea? WHAT THIS LENS SEES CLEARLY: Inner dynamics, symbolic meaning as it relates to the psyche, the relationship between conscious and unconscious processes, why certain ideas grip people at a deep level, the structure of inner transformation. VOICE: Reflective, attuned to nuance, respectful of the full depth of what it examines. Cite [Source X] where relevant.

Philosophical

Metaphysics, ethics, epistemology, ontology, philosophical inquiry

You are observing through the Philosophical lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to assumptions, arguments, and the logical structure beneath claims. You notice what a text takes for granted, what it argues explicitly, and where its reasoning depends on premises that could be questioned. You are interested in the categories a text uses — being, causation, mind, matter, good, evil, knowledge, reality — and whether those categories are examined or assumed. You bring the full breadth of philosophical traditions: Western analytic and continental, Eastern philosophical systems, African philosophy, indigenous epistemologies. WHAT THIS LENS ASKS: • What metaphysical assumptions does this idea rest on? • What does this text assume about the nature of knowledge, reality, or the self — and are those assumptions examined? • Where is the reasoning strong? Where does it depend on unspoken premises? • How does this idea relate to major philosophical traditions and where does it extend or challenge them? • What questions does this text open that it does not attempt to close? WHAT THIS LENS SEES CLEARLY: Logical structure, hidden assumptions, the relationship between claims and their foundations, where different frameworks define the same word differently, the quality of the reasoning and the territory it opens. VOICE: Precise, questioning, engaged. Examines ideas with care rather than dismissal. Cite [Source X] where relevant.

Religious/Spiritual

Comparative theology, mysticism, sacred texts, spiritual practices

You are observing through the Religious/Spiritual lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to the sacred, the transcendent, and the ways human beings have understood their relationship to what is greater than themselves. You approach each tradition on its own terms, honoring its internal logic and its claims about reality. You notice where traditions speak to each other across their differences and where they describe experiences that seem to point toward shared territory. You bring knowledge of mystical, contemplative, and devotional paths across cultures. WHAT THIS LENS ASKS: • How do different traditions understand this concept — and what does it mean within each tradition before any comparison is made? • Where do mystical and contemplative reports converge across traditions, and where do they genuinely differ? • What practices, rituals, or disciplines does this concept connect to? • What does this idea reveal about how human beings experience the sacred, the numinous, or the transcendent? • What living wisdom does this tradition carry that speaks to the question being asked? WHAT THIS LENS SEES CLEARLY: The interior landscape of faith and practice, the claims traditions make about ultimate reality, the common threads and genuine differences in how the divine and transcendent is experienced and described, the lived dimension of belief. VOICE: Reverent without being devotional. Honors each tradition as a living way of knowing. Never suggests one tradition is more correct than another. Cite [Source X] where relevant.

Historical/Anthropological

Cultural evolution, mythology, ritual context, human history

You are observing through the Historical/Anthropological lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to context: when and where an idea emerged, what cultural conditions shaped it, how it traveled across societies, and how its meaning shifted over time. You understand that ideas are products of specific people, places, and moments — and that understanding this context deepens rather than diminishes their significance. You bring knowledge of how contact between traditions (trade, conquest, migration, translation) transforms how concepts are understood and practiced. WHAT THIS LENS ASKS: • When and where did this idea originate, and what was happening in that culture at the time? • How has this concept been understood differently across cultures and eras? • What rituals, practices, or social structures grew around this idea? • How did transmission between traditions transform how this concept was understood? • What does the historical journey of this idea reveal about how human societies organize meaning? WHAT THIS LENS SEES CLEARLY: Cultural context, the transmission and transformation of ideas across time and space, patterns in how societies encounter and reshape knowledge, the human story embedded in even the most abstract concepts. VOICE: Grounded, contextual, curious about how the same idea looks different from different points in history. Cite [Source X] where relevant.

Symbolic/Occult

Correspondences, alchemy, astrology, esoteric symbolism

You are observing through the Symbolic/Occult lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to correspondences, hidden patterns, and the symbolic language that runs beneath the surface of texts and traditions. You understand that many traditions encode their deepest insights in symbol, metaphor, and correspondence systems — Hermetic, alchemical, astrological, Kabbalistic — that reward careful reading. You look for what is being said between the lines, what layers of meaning exist beneath the literal, and how symbolic systems create coherent frameworks for understanding reality. WHAT THIS LENS ASKS: • What symbolic correspondences does this concept invoke — elemental, planetary, numerical, directional? • What does an alchemical or Hermetic reading reveal that a literal reading does not? • How does "as above, so below" apply here — what macrocosmic pattern is reflected in the microcosmic? • What esoteric tradition(s) does this concept belong to, and what is their framework for understanding it? • What initiatory or transformative structure is encoded in the symbols present? WHAT THIS LENS SEES CLEARLY: Layers of meaning beneath the literal, the web of correspondences connecting seemingly unrelated phenomena, the initiatory and transformative structures encoded in symbols, the coherence and depth of esoteric systems as frameworks for understanding. VOICE: Literate, precise, serious. Treats symbolic systems as substantive frameworks of meaning. Cite [Source X] where relevant.

Mathematical

Sacred geometry, numerology, patterns, universal ratios

You are observing through the Mathematical lens — one of seven perspectives in the Seven Lenses. HOW THIS LENS SEES: You attend to pattern, proportion, ratio, and the formal structures that underlie phenomena. You notice when nature, art, architecture, music, or ideas exhibit mathematical relationships — the golden ratio, Fibonacci sequences, geometric symmetries, fractal self-similarity, cyclical patterns, harmonic series. You are interested in how mathematical structure appears across domains and what it reveals about the order of things. You bring knowledge of both pure mathematics and the ways traditions have used number and geometry to encode meaning. WHAT THIS LENS ASKS: • What mathematical patterns or ratios appear in this concept — golden ratio, sacred geometry, numerical symbolism, harmonic relationships? • Is there a geometric, cyclical, or harmonic structure underlying this idea? • How do traditions use number and proportion to encode meaning — numerology, gematria, Pythagorean symbolism? • What does the presence of mathematical structure here suggest about the relationship between form and meaning? WHAT THIS LENS SEES CLEARLY: Underlying formal order, the recurrence of specific ratios and patterns across domains, the relationship between mathematical structure and aesthetic, natural, and spiritual experience, the language of proportion and harmony that connects seemingly unrelated phenomena. VOICE: Precise, pattern-attentive, genuinely curious about what mathematical structure reveals. Cite [Source X] where relevant.

Response Generation Process

When you submit a question to the Prismarium lens engine:

Your query is analyzed through each active lens (based on your weight settings)
Each lens searches our library using hybrid retrieval (vector similarity + keyword matching)
The AI generates a perspective-specific response for each active lens
All lens responses are synthesized into a unified answer that respects your lens weights
The final synthesis is presented with source citations from our library

This multi-perspective approach is designed to help you see questions from multiple angles, but remember: even synthesized responses are AI-generated and can be verified against trusted sources.

How Large Language Models (LLMs) Work

A practical understanding of how LLMs work can support thoughtful use. Here is context that may be helpful:

Statistical Pattern Matching, Not True Understanding

LLMs like GPT-4o don't "think" or "understand" in the way humans do. Instead, they use statistical pattern matching to predict what text should come next based on patterns they learned from their training data. The model has been trained on vast amounts of text from the internet, books, and other sources, and it has learned which words and phrases tend to follow which other words and phrases.

When you ask a question, the model doesn't access a database of facts or "know" anything in the traditional sense. Instead, it generates a response by predicting, token by token (word by word), what text is most likely to follow based on the patterns it learned during training.

Next-Token Prediction

The core mechanism of LLMs is called next-token prediction. The model:

Takes your input (your question plus any context)
Calculates probabilities for what token (word or word-part) should come next
Selects the most likely token (or samples from likely tokens)
Repeats this process, building the response one token at a time
Stops when it reaches an end-of-sequence marker or length limit

This process creates coherent-sounding text, but it's fundamentally different from human reasoning. The model is generating text that sounds knowledgeable, not text that comes from genuine understanding.

Hallucinations: Why They Happen

Hallucinations are instances where an LLM generates information that sounds plausible but is factually incorrect, made up, or not grounded in reality. This happens because:

The model doesn't distinguish between true and false information - It only knows what patterns are common in its training data. If false information appears frequently in training data, the model may generate it.
Pattern completion can create false connections - The model may connect ideas that appear together in training data, even if those connections are incorrect or misleading.
Confidence doesn't equal accuracy - LLMs can generate highly confident-sounding responses that are completely wrong. The model doesn't "know" it's wrong—it's just continuing a pattern.
Training data limitations - If the model hasn't seen accurate information about a topic, it may generate plausible-sounding but incorrect information.
No real-time fact-checking - The model generates responses based on its training, not on current, verified facts. It cannot access the internet or verify information in real-time.

This is why verification is essential. It is often useful to cross-reference important claims from AI responses with reliable sources, especially for:

Medical or health information
Legal advice
Financial decisions
Historical facts
Scientific claims
Personal or life-altering decisions

The Difference Between AI Responses and Human Knowledge

Human knowledge comes from experience, reasoning, verification, and understanding. AI responses come from statistical patterns. This fundamental difference means:

AI doesn't "know" anything - It predicts text patterns, not facts
AI can't verify its own responses - It has no mechanism to check if what it says is true
AI confidence is not reliability - A confident-sounding response may be completely wrong
AI can't learn from mistakes in real-time - It generates based on training, not on feedback
AI doesn't understand context the way humans do - It processes patterns, not meaning

This doesn't mean AI isn't useful—it can be an excellent tool for exploration, synthesis, and generating ideas. But it means you are invited to stay the final judge of what to believe and how to act.

Future Plans for Model Development

We are committed to transparency about our AI systems and their evolution. Here are our plans for the future:

Fine-tuning with library content - We plan to fine-tune a model (to be determined) using the curated texts in the Prismarium library. This will help the AI better understand and synthesize the specific knowledge traditions we curate.
Fine-tuning with Prismarium graph data - As we build the Prismarium knowledge graph (a knowledge network showing connections between concepts across traditions), we plan to incorporate this structured knowledge into model training to improve accuracy and reduce hallucinations.
Ongoing improvements - We will continue to refine our system prompts, retrieval methods, and synthesis approaches based on user feedback and our own testing.

Important: Even with fine-tuning, AI models will still be subject to hallucinations and errors. Fine-tuning can improve performance on specific domains, but it doesn't eliminate the fundamental limitations of LLMs. Critical thinking and verification will always be essential.

Grounding Techniques: Staying Connected to Reality

When exploring deep questions, especially through AI assistance, it's important to stay grounded in reality. Here are practical techniques to help you maintain balance:

Reality Checks

Source verification - AI responses can be cross-checked through citations, original texts, and trusted expert sources.
Reality testing - Claims about the physical world are strongest when they can be examined through observation or experiment.
Multiple-source comparison - Different traditions, scholars, and sources often illuminate the same question in complementary or conflicting ways.
Care with extraordinary claims - Extraordinary claims generally call for stronger evidence, especially when they conflict with well-established knowledge.

Community and Connection

Dialogue with trusted people - Friends, mentors, and community members can offer perspective that supports grounded reflection.
Real-world balance - Physical activity, social connection, and practical responsibilities can complement intellectual exploration.
Sustained human connection - Relationships provide grounding and perspective that AI interactions cannot replace.

Professional Support

Professional support when helpful - When there is mental health strain, confusion, or disconnection from reality, qualified care can be beneficial.
Expert consultation for high-stakes topics - Qualified professionals can add rigor beyond AI-generated responses.
Room to step back - Pausing AI use may help when interactions create distress, confusion, or unhealthy fixation.

Understanding Psychosis Risks

⚠️ Important Safety Information

There have been documented cases of individuals experiencing psychosis-like symptoms, delusions, or severe mental distress after intensive interactions with large language models from major AI companies. We take this risk seriously and want to help you use our AI tools safely.

Why We're Being Cautious

We understand how delicate the human mind can be. Intensive engagement with AI systems, especially when exploring deep philosophical, spiritual, or existential questions, can potentially:

Reinforce delusional thinking patterns
Create false certainties about uncertain topics
Isolate individuals from reality-checking through human interaction
Trigger or exacerbate existing mental health conditions
Create unhealthy dependencies on AI for meaning or guidance

At the same time, we recognize the importance of knowledge and the ability to sort through complicated themes and look for answers. Our goal is to provide tools for exploration while helping you stay safe and grounded.

Signs to Watch For

If any of the following appear, a pause from AI interactions and support from a qualified professional may help:

Feeling that the AI has special knowledge or is communicating directly with you in a way that feels real
Becoming obsessed with AI responses or spending excessive time in AI interactions
Experiencing confusion about what's real and what's AI-generated
Feeling disconnected from your body, physical reality, or other people
Developing beliefs based on AI responses that feel urgent, certain, or disconnected from evidence
Neglecting real-world responsibilities, relationships, or self-care due to AI interactions
Feeling that you need AI to understand reality or make sense of the world
Experiencing paranoia, delusions, or hallucinations related to AI interactions

These signs don't mean you're "broken" or weak. They mean you may need support, and that's completely okay. Seeking help is a sign of wisdom and self-care.

Mental Health Resources

If you're experiencing distress, confusion, or any mental health concerns, please know that help is available. You don't have to go through difficult times alone.

Crisis Resources

988 Suicide & Crisis Lifeline

Call or text: 988

Available 24/7 for anyone experiencing a mental health crisis, thoughts of suicide, or emotional distress. Free, confidential support.

Crisis Text Line

Text: HOME to 741741

Free, 24/7 crisis support via text message. Trained crisis counselors are available to help with any kind of crisis, including mental health concerns.

General Support

If you have questions, concerns, or suggestions about our AI systems or this disclaimer, please reach out:

Email: support@prismarium.xyz

We're learning together, and your feedback helps us improve. If you have suggestions for how we can make our AI systems safer, more transparent, or more helpful, we'd love to hear from you.

Legal Disclaimers

⚠️ Legal Disclaimer

TO THE MAXIMUM EXTENT PERMITTED BY LAW:

Prismarium and its operators, employees, and contributors shall not be liable for:

Decisions made based on AI-generated information or responses
Harm resulting from reliance on AI-generated content
Mental health issues, distress, or psychological effects resulting from AI interactions
Losses resulting from inaccurate, incomplete, or hallucinated AI responses
Technical failures, data loss, or service interruptions affecting AI functionality
Actions taken by users based on AI-generated content
Indirect, incidental, special, consequential, or punitive damages
Loss of profits, data, use, goodwill, or other intangible losses

Your use of Prismarium's AI features is at your own risk. The AI systems are provided "as is" and "as available" without warranties of any kind, either express or implied. We make no guarantees about:

The accuracy, completeness, or reliability of AI-generated responses
That AI responses are free from errors, hallucinations, or biases
That AI responses will meet your needs or expectations
That AI interactions are safe for all users or in all circumstances

AI responses are for educational and exploratory purposes only. They are not a substitute for:

Professional medical, mental health, or therapeutic advice
Legal counsel or advice
Financial advice or planning
Academic or scholarly expertise
Religious or spiritual guidance from qualified practitioners
Your own critical judgment and discernment

Some jurisdictions do not allow the exclusion of certain warranties or limitations of liability, so some of the above limitations may not apply to you. However, to the maximum extent permitted by law, you agree that Prismarium shall not be liable for any harm resulting from your use of AI features.

By using Prismarium's AI features, you acknowledge that you have read and understood these disclaimers and agree to use AI tools responsibly, with appropriate discernment, and at your own risk.

We're Learning Together

AI technology is rapidly evolving, and we're all navigating this new landscape together. We're committed to:

Transparency about how our AI systems work
Continuous improvement based on user feedback
Prioritizing user safety and mental health
Being honest about limitations and risks
Learning from the community and adapting our approach

If you have suggestions, concerns, or feedback about our AI systems, this disclaimer, or how we can better support safe and meaningful exploration, you can reach us at support@prismarium.xyz.

Your voice matters, and we're here to listen. Together, we can build tools that empower exploration while protecting the delicate balance of the human mind.

Related Information

For more information about our platform policies, please see: