What did Enron know? The Enron Brain
The Enron Brain turns Enron email into dated, source-cited pages about the organizations and internal groups it dealt with. It is a this+that prototype for teams exploring how to make changing company knowledge easier to inspect.
Who it's for
- Operations leaders evaluating ways to organize company correspondence that does not fit into structured system-of-record fields.
- Executives and researchers who need to trace what a company knew about counterparties or regulators at a particular time.
- Teams assessing whether cited, time-aware knowledge pages could help them review the history behind company relationships.
Not the right fit if…
- Teams seeking a ready-made brain built from their own live company data: the page describes this as a prototype and demonstrates it using Enron's public email corpus.
How it fits your workflow
The demonstration takes 45,047 Enron emails as its source. An AI identifies entities and builds pages from the correspondence; a person chooses which kinds of entities should receive pages.
Readers open an entity page, inspect dated facts and their cited emails, and use the date slider to see what was known at different points. Facts expire when the source mail indicates they stopped being true.
Pricing
The vendor doesn't publish prices. Check with What did Enron know? The Enron Brain directly.
Prices checked on Oct 24, 2026 from the vendor's site. They can change; confirm before you buy.
Key features
- Source-cited facts
- Facts on the pages cite the email they came from. The page says facts were discarded if their quoted sentence could not be found word for word in the source message.
- Time-aware entity pages
- Pages record dated facts and expire facts when the mail indicates they stopped being true. A slider lets readers view what was known on a selected date.
- Entity discovery and organization
- The pipeline proposed entity types and found organizations and internal Enron entities. A person selected which kinds of entities would receive pages.
- Fact audit
- The page reports that a second AI audited a held-out sample: 93% of sampled facts were fully supported by their email, and none of 500 sampled facts were wrong at their core.
Limitations to know
- The page describes the Enron Brain as a prototype of a product this+that is building.
- This demonstration is based on Enron's public email corpus and eight mailboxes, rather than a live deployment using another company's correspondence.
- For the demonstration, pages were created for organizations mentioned in at least 25 messages and for five selected parts of Enron; the page notes that a real company brain would require different choices about what deserves a page.
Alternatives to consider
-
Context Engine
Choose it if you need shared memory that connected tools can update and retrieve across projects, rather than an email-corpus prototype with dated, cited entity pages.
-
Echobase
Choose it if you want teams to query, create, and analyze an organization's files using deployed AI agents, rather than explore a prototype built from Enron emails.
Getting started
- Open the Enron Brain page and select an organization from the list of starting points.
- Review the entity page and its cited source emails.
- Move the date slider to inspect the page as it appeared at a different point in time.