Frequently Asked Questions

What is Intraspexion®?

Intraspexion is a startup organized as a Delaware for-profit corporation, with headquarters near Seattle. Intraspexion currently offers a patented Deep Learning software system to identify litigation (and other) risks.

What is “Deep Learning”?

Artificial Intelligence (AI) a broad term. A sub-category of AI is Machine Learning (ML). A sub-category of ML is Neural Networks. A sub-category of Neural Networks is Deep Learning (formally a “multi-layer” Neural Network). As the result of academic breakthroughs beginning in 2013, such Deep Learning neural networks have enabled computers “to learn” from either images or words or both. 

Take driverless vehicles and images: after training runs consisting millions of miles, a Deep Learning system can "understand" the camera and Lidar images well enough to accurately control the gas pedal, brakes, and steering wheel.

How does Intraspexion use Deep Learning?

Please see the discussion at the Tab for How It Works. We discussed how we use “deep learning” there. 

What’s the value of identifying a True Positive?

There are two “values.” First, a True Positive is an impetus to begin an internal investigation. Then, law department personnel, perhaps with the assistance of outside counsel, may use email threading and/or access internal databases to decide whether a potential lawsuit is brewing. Then they can advise a control group executive and (hopefully) nip the risk in the bud. Why find the “smoking guns” after the lawsuits are filed? 

What’s the second value?

The second value is that, when True Positives are identified, a Deep Learning system “learns” to be an even better (more accurate) filter. Our system enables human reviewers to tag and store both True Positives and False Positives in a database. When there’s a sufficient amount, we add them to the training set, and the “engine” becomes more specific to the company. 

What’s the first way the training becomes company-specific?

Our “generic” training set consists of factual allegations from a certain number of previously filed lawsuits without regard to the identity of the defendant. For each prospective client, we also extract the factual allegations in previously filed lawsuits against that prospective client for that category of risk and add them to the generic training.  

You mentioned patents. Where does Intraspexion stand with them?

Intraspexion owns eight (8) patents that use Deep Learning. Please see the Tab for 8 Patents for further details.

Currently, for “deep Learning” patents issued from January 1, 2013 through November 13, 2018, the Top 5 patent owners (“Assignees") are IBM (26), Google (25), Microsoft (17), Siemens Healthcare (15), and Intraspexion (8). After Intraspexion, Amazon has 7, and NEC and Nuance are tied with 6 each. To see the spreadsheet, ask for it via the form at Contact Us.

What are the key differences between what Intraspexion does and the tools used in eDiscovery?

1. With eDiscovery, the personnel in corporate law departments are forced to look backward for custodians of documents that are potentially relevant to an already-filed lawsuit. The responsive documents and timeframes are limited to the allegations and claims in the lawsuit. Intraspexion scans emails from every yesterday, so the scope is limited only by the number of specific risks for which an enterprise wants an “early warning.”

2. In eDiscovery, practitioners are accountable to the opposing parties and to the court. This is the “defensibility” issue. Intraspexion is preventive in nature, so corporate law personnel are accountable only to the company for which they work.    

3. In general, eDiscovery is reactive by definition. Intraspexion enables the corporate law department to be proactive.

Aren’t there other ways internal investigations can be triggered?

Yes. Someone on the way out might say to HR personnel, “You’ll be hearing from my attorney.” Or the company may have a “tip line.”

Do you have Office 365 connectivity?

Yes.

How accurate is your system?

We can only tell you about our results using the publicly available Enron dataset. Initially we trained the Deep Learning Engine to "understand" English. Then we added 10,000 Enron non-discrimination emails to the unrelated set, so the model could “understand” English in the context of emails.

Then, as a proxy for the internal company emails "from yesterday,” we processed a held-out set of 20,401 Enron emails.

Result: Our Employment Discrimination "model" called out 25 emails as being "related" to discrimination. Upon human review, 21 emails were marked as False Positives, but four (4) were recognized as True Positives.

The breakthrough was the system's ability to filter 25 emails out of 20,401 emails and report them as "related" to Employment Discrimination, a fraction equal to about  0.001225, which is about one-eighth of one percent. Our System said “never mind” as to about 99.88% of the emails it processed and "take a look" as to about 0.0012%.

Does Intraspexion store any company emails?

No.

Does Intraspexion forward any company emails to anyone other than the company?

No.

Besides "employment discrimination," what other litigation threats (i.e., types of cases) does your system detect?

We’re an early stage startup and we followed the common advice to early stage companies: “Don’t boil the ocean.” So, currently, our system provides insight only to "employment discrimination" risks. As litigators familiar with PACER already know, there are many business-relevant litigation categories there. We can train a Deep Learning model for any such category. We’ll scale out based on customer demand.

Do you offer a cloud solution?

Intraspexion was recently accepted into Microsoft for Startups. We are exploring a "cloud" solution with Azure.

How secure is your system?

For Intraspexion’s “on-prem” solution, emails are turned into number strings behind your firewall, and the words never go beyond it. When we offer a cloud solution as well, the security may be as good or even better.

What Technical assistance do you provide?

For “on-prem” installations, we provide step-by-step instructions. The instructions include both text and images.

What's your pricing structure?

Each prospective customer has a unique litigation profile. Their prevention priorities differ. As a result, we have no set pricing structure. We customize our pricing and we keep our relationship with each customer confidential.

Hmm. What about the privacy concerns of employees?

First, we don't monitor personal devices. However, where a “computer technology resource" (CTR) policy is in place, and the employee is aware of the CTR policy, and agrees to it, it's generally settled that an employee has no reasonable expectation of privacy when a company monitors its own company-owned computers. See Holmes v. Petrovich Development, LLC, 191 CAl.App.4th 1047, 119 Cal.Rptr.3d 878 (2011) (holding that not even the attorney-client privilege can be raised against a company when a CTP policy is in place and the company monitors the emails in its own computer resources).

For the sake of customer confidence and transparency, can you tell us anything about the software you use?

Yes. We are proud to say that our indexing and visualization capabilities are powered by dtSearch®, www.dtsearch.com. We also use Global Vectors for Word Representation (GloVe) and Google's TensorFlow. Both are “open-source.” GloVe and TensorFlow are included in our system under an Apache 2.0 license, the terms of which are here and are incorporated herein in full by this reference. TensorFlow, the TensorFlow logo and any related marks are trademarks of Google Inc.