Editorial Standards

Our editorial mission

Help any reader — from a curious beginner to a professional outside the AI industry — develop an accurate, confident understanding of how AI works, what it can and cannot do, and how it affects their work and daily life. We favor clarity over hype, evidence over speculation, and concrete examples over jargon.

Independence and funding

AI Understanding is a 501(c)(3) nonprofit. Our curriculum is 100% free for learners, but donations are not our only income. We are supported by reader donations, organizational and foundation sponsorships of free educational programs, a non-refundable tool listing submission and review fee, an optional certification verification fee, and a small number of clearly disclosed affiliate relationships.

None of that buys coverage. The listing fee pays for editorial review only and never guarantees approval, a ranking position, or a favorable write-up — rejected submissions are refunded. Certification fees cover credential verification; every quiz and guide behind a certificate is free to take without paying. Our editorial decisions are not influenced by sponsors, advertisers, or affiliate partners, and we do not accept payment in exchange for favorable coverage, story placement, ranking position in our tools directory, or inclusion in any guide or comparison. Our full disclosure policy is at our earnings disclosure page.

How we source and verify information

• Primary sources first. For news, we link to original company announcements, government filings, academic papers, or court documents whenever possible rather than re-reporting other outlets' coverage.
• Two-source rule for contested claims. If a claim is disputed, novel, or based on a single anonymous source, we either confirm it independently or label it clearly as a single-source report.
• Date and version specificity. AI models, prices, and capabilities change frequently. Every article identifies the date of publication and, where relevant, the specific model versions or release dates referenced.
• No fabricated quotes or examples. Quotes are taken from real, attributable sources. Illustrative examples are clearly framed as examples.

Use of AI in our work

We cover AI, and we use AI tools for source discovery, research assistance, summarization, drafting, and quality checks. Our automated news desk can publish without individual human review; those stories use an organizational byline and are distributed as AI-assisted content. Before publication, the system requires an authoritative source and applies relevance, duplication, attribution, image, and formatting checks. Human review is claimed only when a named editor actually performs it. AI-generated illustrations are identified in the article's image credit or provenance information.

Tool reviews and comparisons

Our tool directory and comparison pages exist to help readers choose AI tools. Our review process:

• Tools are included based on relevance and merit, not on whether the vendor pays us.
• Where we earn a commission from a sign-up link, that relationship is disclosed both on the page and on our earnings disclosure.
• Tool comparisons use listed pricing, audience fit, product positioning, and practical workflow notes instead of unsourced star ratings or review counts.
• Vendors cannot pay to change ranking position, recommendation wording, or coverage. They can request factual corrections through our corrections process.

News coverage

Our news section reports on AI industry developments, policy, research, and safety. We prioritize primary-source confirmation, clarity, and context. Stories that do not clear the automated evidence and quality gates are held or rejected. Opinion or analysis pieces are labeled as such, and material developments update an existing canonical story when appropriate instead of creating near-duplicates.

Conflicts of interest

Writers and editors disclose any financial or personal relationship with an organization they are covering. Where a conflict cannot be avoided, the piece is reassigned or the conflict is disclosed prominently in the article.

Comments, contact, and feedback

We welcome reader feedback and corrections. Reach our editorial team at [email protected] or through our contact page. To report a factual error in a published piece, see our corrections policy.

Accessibility and translation

We publish in plain English and translate selected articles for non-English-speaking readers. We aim for WCAG 2.2 AA accessibility on every page, including high-contrast typography, descriptive alt text on images, and keyboard-navigable interactive elements.

This policy is reviewed at least annually. Last reviewed: July 2026.