One of the eleven dimensions behind the Cyborg Score (1-10).
Written by AskCyborg Research. Last reviewed 2026-09-03.
Whether the technology is a durable advantage or a current implementation — technical defensibility, shipping velocity, intellectual property, and the leverage the platform actually provides.
Each dimension is argued before it is scored. A panel of AI analysts takes opposing positions on the material claims in that dimension — the case for, the case against, and the risks neither side raised first — and each claim is challenged, defended, and either survives or is rewritten. The dimension score reflects what survived, weighted by the confidence rating attached to the surviving evidence. The overall Cyborg Score is the synthesis across all eleven, so a single strong dimension does not carry a company and a single weak one does not sink it. The same company put through the same framework produces the same score.
The score rises when the advantage is something a well-funded competitor could not simply rebuild, and when shipping velocity is evidenced rather than claimed. It falls when the technology is a commodity capability described in proprietary language, when patents are counted rather than read, and when the platform leverage is asserted without a second product using it.
Every material claim carries a confidence rating, and the rating is about evidence, not about tone. High confidence means the claim is supported by disclosed, checkable material and survived a challenge from the opposing analyst. Medium means it held up but rests on partial or single-source evidence, and the caveat travels with it. Low means the claim is directionally supported and could not be confirmed — it is kept because omitting it would be a different kind of dishonesty, and it is labelled so it is never read as settled. A dimension scored on low-confidence evidence gets a middling score rather than a strong one: thin evidence is scored as thin evidence, not as bad news and not as good news.
Technical defensibility is assessed from outside, without access to the codebase, the architecture, or the operational reality. A company with a real advantage that has never had to describe it publicly is under-evidenced here, and this dimension will say so rather than infer strength.
A 3 reads like: the differentiator is a widely available capability with a product name attached, patent count is offered as evidence of a moat, and the platform is claimed to be reusable while exactly one product runs on it.
This example is generic and describes no real company. It is a worked illustration of the reasoning, not a rating of anyone.