Which AI model searches the web better on a Japan cold-outreach rules question?
We gave three AI models the same question. The saved records show web search on for GPT-6.1 Sol at High, Qwen3.8-Max, and K3, and do not say for GPT-6.1 Sol at Light and Ultra. This page counts the websites each answer cites and sets that against the answer's length. We have not yet opened the cited links to see whether they back the answers. It compares the models, not their advice.
As asked, word for word.
What's the cold outreach regulations in Japan? and how do domestic companies deal with it?
The question has two parts, the rules and what Japanese companies do about them. The rules sit in statutes and regulator guidance. Company practice would show up in practitioner pages. It tests whether a model finds both kinds of page.
Which model researched further?
Show as a table
| Answer | Words | Websites cited |
|---|
More websites means the model looked further. It does not mean the answer is right. That needs the link check below, which has not run for this question.
Do the cited pages back what the models wrote? Not checked yet.
In a checked comparison, we open every cited link and compare the page with the passage that cites it. For this question that has not happened. The numbers below say how much there is to check.
Cited links · one square per link, none opened yet
Until then, treat every figure in these answers as a lead, not a fact. Open the page before a number goes into your plan. To see what a finished check looks like, read the comparison How effective are interpreters and translation tools at trade-show booths in Japan?.
What this means when you pick a model for research.
Higher effort cited more websites in fewer words.
GPT-6.1 Sol at Light wrote 779 words and cites 3 websites. At High it wrote 744 words and cites 9. At Ultra it wrote 661 words and cites 11.
The Light answer cites only a law database, a regulator, and an association.
Its three websites are the e-Gov law database, the Personal Information Protection Commission, and the Japan Data Communications Association. The High and Ultra answers add company and press-release pages, such as prtimes.jp, support.haihaimail.jp, and legal.hubspot.com.
K3 describes the rules and links nothing.
K3 wrote 758 words on the email and telephone rules and the penalties. The library counts 0 links and 0 websites for it.
Every answer, with its numbers.
| Model | Setting | Asked | Words | Websites | Links |
|---|---|---|---|---|---|
| GPT-6.1 Sol | Light | 5 Oct 2026 | 779 | 3 | 3 |
| GPT-6.1 Sol | High | 5 Oct 2026 | 744 | 9 | 10 |
| GPT-6.1 Sol | Ultra | 5 Oct 2026 | 661 | 11 | 12 |
| Qwen3.8-Max | Thinking | 5 Oct 2026 | 508 | 9 | 10 |
| K3 | High | 5 Oct 2026 | 758 | – |
More model comparisons
- Model comparisonJapan izakaya customer complaints and operator struggles
- Model comparisonWhat sourcing pains does Kato Sangyo face?
- Model comparisonHow effective are interpreters and translation tools at trade-show booths
- Model comparisonMaruto Mizutani: customer segments, strengths, weaknesses, and issues requiring outside help