REVsign.
Litigation

Outlet headline

US trade regulator and 22 states accuse Amazon of taking $20bn with secret surcharges - The Guardian

Headline trimmed from the source, unedited

Amazon · ecommerce_logistics · 2026-09-09
The US’s main trade regulator and 22 states sued Amazon on Monday, alleging the online retailer “secretly and systematically overcharged” advertisers on Amazon.com.
Verbatim quote · www.theguardian.com ↗read on 9/9/2026 · fingerprint 57c0ff64…

Read original source ↗

Who read this fact

Each model reads the same question separately: whether the quote from the source says what the fact states. A person reads when the models have no clear majority.

Read by GLM-4-9B Not enough to decide

An open-weights model we deploy ourselves. It only accepts what the quote states plainly, nothing that has to be assumed.

Details
Model
GLM-4-9B
Latency
675 ms
Date
Sep 9, 2026, 10:52 PM
Instruction
fcef78fa40ff…
Input evaluated
4b8e853ceb65…
Read by Llama 3.1 8B Supports the fact

An open-weights model we deploy ourselves. It only accepts what the quote states plainly, nothing that has to be assumed.

Details
Model
Llama 3.1 8B
Latency
1401 ms
Date
Oct 18, 2026, 5:53 PM
Instruction
fcef78fa40ff…
Input evaluated
97eb6b9368ba…
Read by CLM-8B Does not support it

An open-weights model that does not write. It only answers whether the quote supports the fact or not.

Details
Model
CLM-8B
Latency
389 ms
Date
Nov 11, 2026, 4:15 PM
Input evaluated
97eb6b9368ba…
Read by Claude Haiku 4.5 Not enough to decide

A model from an outside provider, with the same instruction and without seeing the other readings.

Details
Model
Claude Haiku 4.5
Latency
777 ms
Date
Sep 9, 2026, 10:22 PM
Instruction
fcef78fa40ff…
Input evaluated
4b8e853ceb65…
Read by Claude Sonnet 5 Not enough to decide

A model from an outside provider, with the same instruction and without seeing the other readings.

Details
Model
Claude Sonnet 5
Latency
1210 ms
Date
Nov 1, 2026, 8:48 PM
Instruction
fcef78fa40ff…
Input evaluated
97eb6b9368ba…
Read by Muse Ember 1.0 Not enough to decide

A model from an outside provider, with the same instruction and without seeing the other readings.

Details
Model
Muse Ember 1.0
Latency
1543 ms
Date
Nov 10, 2026, 10:25 PM
Instruction
fcef78fa40ff…
Input evaluated
97eb6b9368ba…
Read by a person Not reviewed

A person on the team reads the quote without seeing what the models said and signs their reading. When it exists, it decides the badge.

No human reading. The models agreed by a clear majority. A person reads only when there is none.

Company exposure

Lawsuits, claims and court rulings against the company

Primary risk vectorLitigation1

Concurrent risksNone detected.

The thread

It is the only fact in this thread for now.

Why it is here

List of published signals

Every signal cites its source. If it cannot be audited, the data does not exist.
Mandate 03 · Non-Negotiable Traceability →

This signal as data
curl https://revsignlab.com/en/senal/1a114cae-8a11-450b-b92f-818b18f1c5b2/json
MCP query to paste into your assistant
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "obtener_senal",
    "arguments": {
      "id": "1a114cae-8a11-450b-b92f-818b18f1c5b2"
    }
  }
}

← See all radar signals