AI Transparency

1. Overview

unbuilt.me (the “Service”) uses artificial intelligence, automated systems, and third-party model providers to generate analyses, reports, recommendations, summaries, stack suggestions, product guidance, and other outputs.

These outputs are generated automatically in response to user inputs and system instructions.

The Service is designed to assist with exploration and decision support. It is not a source of guaranteed truth, professional advice, or verified factual conclusions.

2. How the System Works

The Service may process and combine:

• user-provided prompts, ideas, and other Inputs,
• system instructions and internal prompt logic,
• model outputs generated by third-party AI providers,
• search or enrichment data from third-party providers,
• internal formatting, gating, scoring, or presentation logic,
• account or plan status that determines which output depth is shown.

Outputs are generated probabilistically and are not deterministic. The same or similar input may produce different outputs at different times.

3. Free vs Paid Output

The Service may provide different output depth depending on plan type, usage status, feature access, or whether an analysis is generated through a free or paid flow.

For example:

• free users may receive limited, reduced-depth, lower-cost, or gated output;
• paid users may receive broader, deeper, or higher-cost output;
• users who exhaust included usage may fall back to limited output modes.

These differences are part of the product design and do not imply that any output is complete, comprehensive, authoritative, or guaranteed.

4. What AI Can and Cannot Do

AI systems can identify patterns, summarize information, generate structured analysis, and produce useful suggestions.

However, AI systems also have important limitations.

AI-generated outputs may be:

• incorrect,
• incomplete,
• inconsistent,
• misleading,
• outdated,
• oversimplified,
• speculative,
• or poorly calibrated for your specific situation.

AI systems may:

• hallucinate facts or sources,
• overstate confidence,
• miss key context,
• misread niche markets,
• reflect biases in training data or retrieved information,
• or produce outputs that sound convincing while still being wrong.

You must not treat AI output as automatically true, verified, complete, or suitable for high-stakes reliance.

5. No Professional Advice

The Service does not provide:

• legal advice,
• financial advice,
• tax advice,
• investment advice,
• accounting advice,
• employment advice,
• medical advice,
• or any other regulated professional advice.

Nothing in the Service should be interpreted as a recommendation to invest, launch, buy, sell, hire, fire, enter into contracts, form a company, choose a legal structure, or take any high-stakes business or legal action without your own independent review and, where appropriate, professional advice.

6. No Guarantee of Accuracy or Results

We do not guarantee that any output is:

• accurate,
• complete,
• current,
• reliable,
• unbiased,
• merchantable,
• fit for a particular purpose,
• or suitable for your business, product, market, or legal/commercial decisions.

We do not guarantee:

• that an idea score is correct,
• that a recommendation is sound,
• that a market gap is real,
• that a competitor assessment is complete,
• that a suggested stack is optimal,
• or that following an output will produce any commercial result.

The Service is not a guarantee engine, predictive certainty system, or substitute for validation in the real world.

7. Third-Party AI and Data Providers

We rely on third-party providers to generate and enrich outputs. These may include, depending on product configuration over time:

• AI/model providers,
• search or enrichment providers,
• hosting and infrastructure providers,
• and related service vendors.

Your inputs may be transmitted to and processed by such providers in order to generate outputs or operate the Service.

We do not control every aspect of how third-party systems generate results. We also do not guarantee uninterrupted availability, consistency, or quality from any third-party provider.

8. Search, Retrieval, and Knowledge Limits

Some outputs may rely on:

• model knowledge,
• search-enriched data,
• retrieved content,
• or no external retrieval at all.

Accordingly:

• some outputs may reflect older model knowledge rather than live data,
• some outputs may be based on incomplete search or retrieval results,
• some outputs may omit relevant competitors, market signals, risks, or context,
• and some outputs may present partial information in a simplified or summarized way.

Even when external data is used, the Service does not guarantee that all relevant sources were found, interpreted correctly, or incorporated into the final result.

9. Scoring and Recommendation Limits

Where the Service provides:

• scores,
• labels,
• recommendations,
• or action suggestions,

those are generated by automated systems using internal logic, scoring rules, and AI interpretation.

Such outputs are:

• informational,
• non-binding,
• and not guarantees of business quality or commercial viability.

A low score does not prove an idea cannot work. A high score does not prove an idea will work. A recommendation is not a guarantee or instruction.

10. Transparency Limits

AI systems and multi-step software systems are inherently complex.

Although we may describe some aspects of how the Service works, we do not guarantee full interpretability, traceability, or explainability for every output.

We may not be able to fully explain:

• why a model chose a particular wording,
• why a result differed from another run,
• why a certain competitor was included or excluded,
• or why a score changed after model, prompt, search, or product updates.

You acknowledge that full output traceability may not be available.

11. Continuous Changes and Improvement

We continuously test, refine, and change the Service.

That may include changes to:

• prompts,
• scoring presentation,
• models,
• routing,
• retrieval logic,
• output structure,
• plan entitlements,
• free vs paid depth,
• and product flows.

As a result:

• outputs may change over time,
• the same idea may receive different wording or structure later,
• and historical outputs may not match future outputs.

Improvement does not guarantee correctness.

12. Your Responsibility

You are solely responsible for:

• evaluating outputs critically,
• verifying important claims independently,
• checking business, legal, financial, and market assumptions yourself,
• testing ideas in the real world,
• and deciding whether or how to act on any output.

You should not rely exclusively on the Service for any important decision.

At a minimum, you should:

• sanity-check material claims,
• validate important facts independently,
• treat outputs as starting points, not final truth,
• and consult qualified professionals where needed.

13. Sensitive and High-Stakes Use

You should not use the Service as the sole basis for:

• legal compliance,
• securities or investment decisions,
• regulated business advice,
• tax strategy,
• material contractual commitments,
• or other high-stakes decisions.

If you choose to use the Service in those contexts, you do so entirely at your own risk.

14. Data Usage Related to AI Processing

To provide AI features, we may process your inputs and outputs through internal systems and third-party providers.

We do not sell your data for third-party advertising.

We may process inputs and outputs to:

• generate results,
• maintain account history or saved reports,
• improve service performance and reliability,
• investigate abuse or misuse,
• debug product issues,
• and enforce our Terms and access rules.

For more detail, see our Privacy Policy.

15. No Duty to Monitor or Correct

We have no obligation to:

• manually review every output,
• fact-check every analysis,
• correct every inaccuracy,
• or monitor how you use outputs.

We may improve the system over time, but we do not undertake a legal duty to supervise, certify, endorse, or correct outputs for your individual use case.

16. Contact

If you have questions about this AI Transparency Policy, contact:

builder@unbuilt.me