Academic Disclaimer

01 Nature of AI Outputs

SciLens uses large language models (LLMs) — including Google Gemini and DeepSeek — to analyze research documents and generate summaries, extractions, annotations, and other outputs. These models are probabilistic systems: they predict statistically likely text, not factual truth.

AI outputs generated by SciLens are best understood as a first-pass reading aid. They can highlight relevant sections, extract surface-level structure, and suggest connections — but they cannot reason with the same rigor as a trained domain expert, and they do not have access to context beyond what you provide.

02 What SciLens Cannot Do

Regardless of how accurate an output appears, SciLens is not a substitute for:

• Peer review or expert editorial judgment;
• Academic supervision by a qualified researcher or advisor;
• Clinical review, medical diagnosis, or treatment recommendations;
• Legal advice, regulatory guidance, or compliance assessment;
• Engineering safety review or technical certification;
• Statistical verification or replication of experimental results;
• Fact-checking against primary sources or authoritative databases.

03 Hallucination Risk

LLMs are known to produce "hallucinations" — outputs that appear plausible but are factually incorrect, fabricated, or internally inconsistent. This risk is present in every SciLens output, regardless of the model used or the quality of the source document.

AxonReady is actively working to improve output accuracy, but no current AI system can guarantee hallucination-free outputs. The responsibility for detecting and correcting errors rests entirely with the user.

04 Citation & Reference Accuracy

SciLens may extract, summarize, or reformat bibliographic information. You must verify:

• Author names, affiliations, and publication years are correct;
• DOIs, journal names, and volume/page numbers are accurate;
• The cited work actually exists and contains the attributed claim;
• Paraphrased quotes accurately represent the source's meaning;
• Citation formatting complies with the required style (APA, MLA, Chicago, Vancouver, etc.).

Never submit AI-extracted citations to academic papers, grant proposals, or publications without verifying each reference against the primary source.

05 Figure, Chart & Data Extraction

SciLens uses vision models (Google Gemini) to interpret figures, charts, tables, and images. These interpretations are approximations. You must verify:

• Numerical values extracted from charts and graphs;
• Units of measurement and axis labels;
• Statistical significance values, confidence intervals, and error bars;
• Color-coded data points, especially in complex visualizations;
• Captions and their relationship to described figures.

Vision model accuracy degrades with low-resolution images, complex multi-panel figures, or non-standard chart types. Always refer to the original figure in the source document.

06 Academic Integrity

Using SciLens outputs in academic work is subject to your institution's academic integrity policies. Policies vary — some institutions prohibit AI-assisted writing entirely; others require disclosure.

SciLens is intended as a reading and comprehension tool — analogous to a search engine or a highlighter. Using it to generate text that you present as your own original analysis may violate academic integrity rules at your institution.

07 Professional & Sensitive Domains

SciLens outputs carry heightened risk when applied to the following domains. Do not make consequential decisions based on SciLens outputs in these fields without expert review:

08 Export & Downstream Use

SciLens allows you to export analyses in various formats. Once exported, you assume full responsibility for how that content is used, shared, or published. AxonReady has no visibility into or control over downstream use of exported outputs.

If you incorporate SciLens-generated content into publications, presentations, or reports, you are responsible for ensuring accuracy, proper attribution, and compliance with your institution's, journal's, or employer's policies.

See the full export liability provisions in our Terms of Service, Section 7 →

09 Your Obligation to Verify

By using SciLens, you accept the obligation to independently verify all outputs before relying on them. Specifically, you agree to:

• Read the source document yourself rather than relying solely on AI summaries;
• Cross-reference extracted citations with original sources;
• Verify numerical data, statistics, and measurements against the original figures;
• Apply domain expertise to evaluate the reasonableness of conclusions;
• Consult qualified professionals in regulated fields before acting on AI-generated insights.

10 No Liability

AxonReady is not liable for any harm — academic, professional, clinical, financial, reputational, or otherwise — arising from your reliance on SciLens outputs. This includes but is not limited to: research errors, incorrect citations, failed publications, academic sanctions, clinical errors, business decisions made on the basis of AI summaries, or any downstream consequences of exported content.

This disclaimer is incorporated into and forms part of the Terms of Service. The limitations of liability and warranty disclaimers in those Terms apply fully to all SciLens academic use.

11 Questions