How Termcon analyzes agreements

Termcon uses a large language model to read the full text of a Terms and Conditions document and evaluate it across four categories. Each category receives a risk level, a plain-language explanation, and a short quote or paraphrase from the text that supports the assessment.

The four categories

Termcon reviews every agreement for: forced arbitration (whether you must resolve disputes privately and waive class actions), data collection and sharing (how broadly your personal data is gathered, used, and disclosed), limitation of liability (how broadly the company caps or excludes its responsibility), and unilateral changes (whether the company can rewrite the terms without meaningful notice).

How risky language is identified

The model reads the document and looks for clauses matching each category. It then assigns a risk level (No Risk, Low Risk, Medium Risk, or High Risk) based on how broadly the clause applies. A narrow, optional clause scores lower; a broad, mandatory clause that removes significant rights scores higher.

How to interpret the results

Each category score follows a fixed scale: No Risk is 0, Low Risk is 1, Medium Risk is 1.75, and High Risk is 2.5. The overall score is the sum of all four (maximum 10), where lower is better for you. The evidence quote shows where in the text the assessment comes from, so you can verify it yourself.

Limitations of automated analysis

Automated analysis is fast but not perfect. The model may miss nuanced language, misread context, or fail to account for laws that vary by jurisdiction. Termcon is meant as a quick first read to spot the major clauses. It's no substitute for a full legal review.

Informational only

Termcon's results are informational and do not constitute legal advice. They are not a substitute for review by a qualified attorney. If a decision has significant legal or financial consequences, consult a lawyer.