Detector AI & Analysis Disclaimer

Super Nova Research Inc., operating as Draft&Goal
Version 1.0 — Last updated July 16, 2026

This AI & Analysis Disclaimer (“Disclaimer”) applies to Detector and forms part of the Detector Terms of Service. Capitalized terms not defined here have the meanings in the Terms.

1. What Detector Does

Detector uses a combination of statistical methods, machine-learning systems, deterministic rules, and, where applicable, generative AI to analyze submitted text, files, URLs, domains, sitemaps, and other Analysis Inputs. Depending on the feature, Output may include AI-authorship classifications, highlighted passages, content-quality observations, technical or search-related findings, website scores, accessibility indicators, AI-search-readiness assessments, explanations, edits, and recommendations.

Detector is a decision-support and informational tool. It does not independently verify facts, authorship, ownership, intent, legal compliance, accessibility compliance, security, or business outcomes.

2. AI-Authorship Detection Is Probabilistic

AI-generated-text detection is an evolving and inherently uncertain field. Human writing may be classified as AI-generated, and AI-generated or AI-assisted writing may be classified as human. Results can vary based on language, length, topic, editing, translation, quotation, writing style, model family, and changes to detection methods.

A score, label, highlighted passage, or statement that text “appears” AI-generated:

• is a statistical assessment, not a statement of fact;
• does not identify which person or system wrote the text;
• does not establish plagiarism, cheating, fraud, deception, copyright infringement, or policy breach;
• does not prove that AI was or was not used; and
• should not be represented as forensic, scientific, or legally conclusive evidence.

Short, highly formulaic, translated, edited, technical, multilingual, or non-native writing may be especially difficult to classify. No detector can reliably identify all AI-generated or AI-assisted content in every context.

3. Website and Content Analysis Is a Snapshot

Website Output reflects information and technical conditions available when an analysis is run. Pages, search engines, AI-answer systems, ranking methods, browser behavior, accessibility requirements, third-party platforms, and industry practices change frequently. Crawling may be incomplete because of authentication, robots controls, rate limits, JavaScript rendering, location, personalization, outages, paywalls, blocked resources, or other technical conditions.

Detector does not guarantee:

• crawling, indexing, ranking, search visibility, citation, or inclusion in AI-generated answers;
• increased traffic, engagement, revenue, conversions, grades, acceptance, or other outcomes;
• conformance with WCAG or any accessibility, privacy, security, advertising, consumer, or sector-specific law;
• that all errors, risks, or opportunities will be identified;
• that third-party platform rules or algorithms are accurately inferred; or
• that implementing a recommendation will improve performance or avoid adverse effects.

Scores are relative indicators produced by proprietary methods. They may change when Detector’s methods or the analyzed material changes and may not be comparable across different versions, languages, sites, or tools.

4. Generative AI Output

Explanations, recommendations, summaries, and suggested edits may be generated in whole or part by AI. Generative AI can produce inaccurate, incomplete, biased, outdated, inconsistent, or fabricated information. It can omit important context and may repeat patterns found in other material.

You must review and validate Output before relying on, publishing, submitting, or implementing it. You are responsible for fact-checking, source verification, originality review, rights clearance, testing, and professional review appropriate to the use.

5. Human Review Is Required

Detector Output must not be the sole basis for a decision that affects a person’s rights, opportunities, reputation, education, employment, access to services, or other significant interests. This includes accusations or findings of academic misconduct, workplace discipline, hiring, admissions, grading, professional licensing, credit, insurance, housing, healthcare, or legal consequences.

Before taking such action, the responsible decision-maker must, at minimum:

1. review the underlying material and relevant context;
2. consider independent, reliable evidence;
3. account for false positives, false negatives, language, disability, and other relevant factors;
4. involve a qualified human with authority to change the outcome;
5. tell the affected person how Detector Output was used where required or appropriate;
6. provide a meaningful opportunity to respond, contest, and correct; and
7. comply with applicable privacy, human-rights, education, employment, consumer, and AI laws.

Detector is not designed or authorized to make autonomous decisions with legal or similarly significant effects.

6. No Professional Advice

Output is not legal, financial, tax, medical, psychological, educational, accessibility, cybersecurity, engineering, or other regulated professional advice. It does not replace an audit, penetration test, accessibility evaluation, legal review, editorial review, or advice from a qualified professional familiar with your circumstances.

No attorney-client, accountant-client, doctor-patient, fiduciary, or other professional relationship is created by using Detector.

7. User Responsibilities

You are responsible for:

• ensuring that you are authorized to submit or analyze the material;
• respecting privacy, confidentiality, copyright, database, contractual, and other rights;
• not submitting credentials, secrets, regulated data, or sensitive Personal Information unless expressly supported and authorized;
• using Output honestly and describing its limitations when sharing it;
• testing recommendations in an appropriate environment before deployment;
• maintaining backups and change-control procedures; and
• complying with the Acceptable Use Policy.

Do not use editing or rewriting features to conceal misconduct, misrepresent authorship, defeat legitimate provenance or integrity controls, or violate an academic, employer, publisher, platform, or professional rule.

8. Third-Party Systems

Detector may discuss or evaluate third-party search engines, AI systems, websites, models, standards, or platforms. Draft&Goal is not affiliated with or endorsed by those parties unless expressly stated. Their products, policies, algorithms, and availability are outside our control and may change without notice.

External links and references are provided for convenience and are governed by third-party terms. Customer is responsible for evaluating them.

9. Research, Benchmarks, and Public Claims

Benchmarks and accuracy statements are meaningful only in relation to the dataset, language, model versions, thresholds, evaluation protocol, and date used. Customer must not publish comparative performance claims based on Detector without a method that is fair, reproducible, statistically appropriate, and compliant with the Terms.

We may describe validation results or confidence measures, but no result eliminates the limitations in this Disclaimer.

10. Reporting Concerns

To report a suspected false positive, false negative, unsafe recommendation, bias concern, or other quality issue, contact detector@dng.ai with the report identifier and enough non-sensitive context to investigate. Do not email confidential Analysis Inputs unless support asks you to use an approved secure channel.

Privacy concerns: privacy@dng.ai
Security concerns: security@dng.ai
Legal matters: legal@dng.ai

11. Relationship to the Terms

The warranty disclaimers and liability limitations in the Terms apply to Detector and Output. This Disclaimer does not exclude rights or remedies that cannot lawfully be excluded, including applicable consumer rights.