Enron.
Defined preprocessed Enron email table: 13,508 records; approximately 17.2 million characters; approximately 2.4 million words; processing time 4 minutes 47 seconds. No internet or later history as input.
Cold Case Studies
EOS LUX — See the unseen.
Method demonstration using public or court-released source material. Not legal advice. Not a finding, allegation, legal opinion, or determination of any kind.
Method statement
Method statement
EOS LUX was applied to defined public or court-released records, not to general internet commentary or retrospective narrative. Historical context is applied after the run as validation, not as input. No outcome, verdict, later reporting, or hindsight was provided to EOS LUX as input at any stage of either run.
EOS LUX can be tested against public or court-released cold-case records because the broader historical outcomes are already known. The point is not that EOS LUX knew the history. It did not. The point is that, from defined records alone, EOS LUX surfaced source anchors, omissions, chronology shifts, behavioural indicators and review questions that align with issues later understood as important.
What this report includes and does not include
What this report includes and does not include
This report includes source-anchored quotes, documented omissions, chronology metadata, behavioural and role indicators drawn only from the record, known limitations, and human follow-up questions — for two defined datasets: the Enron email archive and Exhibit H from the Musk / Twitter litigation.
This report does not explain what EOS LUX is or how it works as a product, and it does not address buyer roles, industries, or comparisons with other tools. It does not determine motive, guilt, liability, authenticity, or admissibility.
Every card ends: not a finding, allegation, legal opinion, or determination.
Source environments
Source environments
Musk / Twitter.
Exhibit H from Twitter, Inc. v. Elon R. Musk et al. — court-released text-message exhibit, not a leak. Stricter legal-sensitivity pass than Enron.
Two-run stability — Enron only
Enron only: the same source anchors and omission flags recurred across two independent runs on that defined dataset. The second run layered additional structure rather than contradicting the first. Do not read this two-run claim as applying to the Musk / Twitter exhibit. Do not publish internal reliability percentages.
What the cold cases prove
What the cold cases prove
From defined records alone, EOS LUX can produce: source anchors; omission / silence signals; chronology; record-derived behavioural / role indicators; provenance notes; human-review questions; a repeatable first map.
They do not prove historical guilt, fraud, or the later public narrative.
What EOS LUX does and does not do
What EOS LUX does and does not do
EOS LUX does not provide legal advice, determine guilt, prove fraud, decide who is lying, diagnose people, determine admissibility, or replace professional judgment. It helps responsible reviewers see what the record contains, what it omits, and what should be examined first. It does not replace lawyers, investigators, HR teams, compliance officers, or boards. It is not a truth detector, an autonomous forensic system, or a psychological profiling tool.
Every output requires qualified human review before any action, allegation or decision relies on it.
CTA
Request a Confidential Briefing.
Request the cold-case studies PDF via a business-email request; no file upload.
Request the cold-case studies PDF