Editorial Policy
Last updated: June 1, 2026
At Work Agent our mission is to provide accurate, unbiased and up-to-date information about AI tools. This page explains our editorial process.
1. Independence
- No AI vendor pays us for scores, placement or favorable coverage.
- We do not use referral/affiliate links that could bias our scoring.
- All revenue comes from display advertising (Google AdSense) which is independent of the tools we review.
2. How We Research
- Hands-on testing: Every model is tested by a named editor using our standard 50-prompt benchmark covering coding, writing, research, math, and multilingual tasks.
- Public benchmarks: We cite official scores (MMLU, HumanEval, GSM8K) where available, sourced from papers, changelogs or the vendor's own documentation.
- Pricing verification: Pricing is checked against the vendor's pricing page on the review date, with a link to the source.
3. Scoring System
Each AI product is rated on a 0–10 scale across four editorial dimensions: Overall, Value for Money, Accuracy and Ease of Use. Additionally, seven capability dimensions are scored 0–10 based on our prompt benchmark.
Scores are set by the reviewing editor and reviewed by a second team member. Disagreements above 1.0 points are resolved by re-testing.
4. Use of AI in Our Workflow
We are transparent about our own use of AI:
- AI tools may assist with initial research drafts and outline generation.
- All final content is written, fact-checked and edited by a human editor.
- Every article undergoes manual review before publication.
- AI-generated images or diagrams are clearly labelled.
5. Corrections & Updates
- We review each product page at least once per month.
- If a vendor ships a material update (new version, pricing change), we aim to update within 7 days.
- Corrections reported by readers are investigated within 24 hours.
- Every article displays the last-updated date.
6. Sources & Citations
We cite at least two external sources per claim where applicable. Source links appear inline or in a “References” section at the bottom of long-form articles.
7. Feedback
If you believe any information is inaccurate, please contact us at Hunk77550@gmail.com with the URL, the claim in question, and a link to a correcting source.