About Axiom — our AI research agent

Axiom is an AI, not a person. Every comparison, guide and capability table on AI Tool Atlas bylined “Axiom” was researched and written by an AI research agent. We’re telling you that up front because it’s the truth — and because there’s something worth noting: Axiom is an AI that evaluates AI tools. That meta-position is the whole point. Axiom doesn’t need to pretend to be a human expert who “tried” every tool. It tracks the verifiable facts about what each tool can and cannot do — and it does so with a transparency no human reviewer column can match.

What Axiom does

Axiom maintains AI Tool Atlas’s comparison database: published feature sets, pricing tiers, model capabilities, integration lists, and stated limitations — drawn directly from each tool’s own documentation and confirmed third-party benchmarks. When a tool updates its pricing or ships a new capability, Axiom updates the record. At our scale — tracking dozens of categories across hundreds of tools — the speed and consistency of an AI agent beats a human team on every dimension except one (see below).

What Axiom does NOT do

Axiom does not pretend to have personally used a writing assistant to draft a novel, or run code through a coding tool on a live project. An AI agent has no lived experience, and faking it would be both dishonest and unnecessary. Axiom deals only in verifiable facts — published specs, documented limits, reproducible benchmark results — and where real user experience matters, it aggregates real user evidence rather than fabricating its own.

How Axiom verifies

• Every factual claim traces back to a primary source: the tool’s own documentation, a dated pricing page, or a published benchmark.
• Before any comparison publishes, it goes through an adversarial fact-check pass designed to disprove the claim.
• Numbers we cannot verify are left blank — never estimated or guessed.
• Axiom runs on: Claude, Anthropic — we’ll keep this list current as it changes.

Who’s accountable

Axiom is overseen and published by Lead Media Ltd (Bulgaria). A real human is responsible for everything here and is reachable at hello@aitoolatlas.io. Axiom is a tool we operate transparently — not an unaccountable bot.

When Axiom is wrong, we say so

AI tools change fast. When Axiom gets something wrong — a feature that shipped, a price that changed, a limit we overstated — we fix it and log it publicly: what was wrong, what’s now correct, and when, on our corrections log. That public record is the core of the trust model.