Intraspexion Inc.

Company Phone:
Company Address:
SEQUIM, WA, 98382-8461
Company Purpose:
Intraspexion is a patented Deep Learning system that enables corporate counsel to identify a risk before it becomes a contested dispute.

It's an early warning system that eliminates or reduces litigation fees & costs through dispute avoidance.

Corporate counsel is closest to their internal communications data but they are blind to the risks. Our patented software uses classified text and Deep Learning algorithms to identify risk and provide an early warning of potential litigation to corporate counsel. This early warning enables them to investigate and advise their control group executives, enabling them to be proactive and nip the litigation risk in the bud.

Intraspexion was founded on the idea that prevention was the best way to deal with the enormous cost of litigation. The cost is on the order of $350,000 to $400,000 per case, even if a company wins in court. So the only way to win is not to play. Before Intraspexion, no one had taken on that challenge. With Deep Learning we experimented, built a team, and put the enterprise-grade software together.


Competitive Advantage:
After we solved the problem, and saw how we could extend it from one use case (employment discrimination) to many others (well over 50), we filed for and earned a patent (No, 9,552,548, issued on January 24, 2017) in less than four (4) months after the application was filed. (Light speed for the patent office.) We have six additional patent applications pending now, and all of them have been granted expedited review.
Favorite Quote:
Gretzky's advice - Go where the puck is going to be.
Summary of Best Use Cases:

Pilot Program. Free* -- but for corporate legal only.

Our starting "use case" is employment discrimination.

We already have Proof of Principle on Enron emails. And, of course, we have no access to prospective customer emails.

So how about your emails? If your company has experienced 100 or more employment discrimination lawsuits in federal court from January 1, 2012 to December 31, 2016, we'll use that publicly available information (which contains zero eDiscovery materials) to re-train our Deep Learning model to be specific to you.

Then, we'll send you a server (loaded with Deep Learning language model specific to you) and a laptop for you to use "on premises" for 21 days after we ship our hardware to you.

You'll ask IT to set the hardware up in a war room and to isolate our server-laptop combination from any other system your company uses. That way, IT should have no objections.

Then you can test our system by putting in (a) some Enron emails (for testing purposes) and (b) then your not-public emails (in .pst format) from the production set(s) from one or more of your now-closed cases.

Can our system find one or more of the risky emails you flagged during the case and already know about?

Can our system find a risky email you didn't know about? (Yes, we've done it before.)

If you answer either question "yes," then you have a Proof of Concept, using your (still-private) data, and then you can ask IT to wipe our server and laptop before you ship them back to us.*

But then you can ask us about pricing and terms.

So ask us today for the free* pilot agreement. Use the Contact Us page. First come, first served.

* means you pay shipping costs only (both ways).




Categories (to select multiple categories hold shift key down):
Algorithms that Learn and Adapt
Deep Learning / Machine Learning (General)
Thought Leader