ads.txt is a transparency layer, not proof of quality
ads.txt records which sellers a publisher declares as direct or reseller. Because the information is declared, the format alone cannot prevent false statements, compromised relationships or manipulated bid requests.
Inclusion lists reduce the attack surface
Fou recommends whitelists or inclusion lists of known sites and apps. They narrow the open inventory universe but do not replace measurement because fraud actors can impersonate allowed domains or apps.
Fewer exchanges simplify the supply chain
In an experiment described by Fou, correct delivery to the intended domain improved materially when a single exchange was also specified. SPO can therefore reduce uncertainty as well as cost.
Compare the bid request with the actual placement
The key forensic question is whether the domain, app and other supply data declared before the bid match the environment in which the creative actually rendered after the bid.
Post-bid evidence exposes discrepancies
An in-ad tag can observe placement and browser information. When declared and observed data diverge, advertisers have a concrete lead for investigating spoofing, leakage or misconfigured supply paths.
This article describes principles and questions. Specific fraud, bot or quality rates cannot be transferred directly to individual campaigns without understanding methodology, scope and data basis.