`
`Transcript of Aviel D. Rubin, Ph.D.
`
`Date: February 17, 2025
`Case: Datavant, Inc. -v- Vigilytics LLC (PTAB)
`
`Planet Depos
`Phone: 888.433.3767 | Email: transcripts@planetdepos.com
`www.planetdepos.com
`Michigan #8598 | Nevada #089F | New Mexico #566
`
`WORLDWIDE COURT REPORTING & LITIGATION TECHNOLOGY
`
`Datavant Exhibit 1025
`Datavant v. Vigilytics, IPR2024-00311
`
`
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
` ______________
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
` ______________
` DATAVANT, INC.,
` Petitioner,
` v.
` VIGILYTICS LLC,
` Patent Owner.
` ______________
` IPR2024-00311, IPR2024-00381, & IPR2024-00382
` Patent Nos. 9,965,651, 10,109,375,
` 10,886,012 and 9,665,685
` ______________
` Deposition of AVIEL D. RUBIN, Ph.D.
` Conducted Virtually
` Monday, February 17, 2025
` 9:30 a.m. EST
`
`Job No.: 571619
`Pages 1 - 93
`Reported by: Debra A. Whitehead
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`2
`
` Deposition of AVIEL D. RUBIN, Ph.D., conducted
`virtually.
`
` Pursuant to notice, before Debra Ann Whitehead,
`E-Notary Public in and for the Commonwealth of
`Virginia.
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`3
`
` A P P E A R A N C E S
` ON BEHALF OF PETITIONER:
` JUSTIN J. OLIVER, ESQUIRE
` VENABLE LLP
` 600 Massachusetts Avenue, NW
` Washington, DC 20001
` (202) 344-4000
`
`ON BEHALF OF PATENT OWNER:
` NICHOLAS W. STEPHENS, ESQUIRE
` FISH & RICHARDSON, PC
` RBC Plaza
` 60 South 6th Street
` Suite 3200
` Minneapolis, Minnesota 55402
` (612) 355-5070
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`4
`
` C O N T E N T S
`EXAMINATION OF AVIEL D. RUBIN, Ph.D. PAGE
` By Mr. Oliver 5
` By Mr. Stephens 73
`
` EXHIBITS MARKED IN TODAY'S SESSION
` (None)
`
` EXHIBITS MARKED IN PRIOR SESSIONS
` (Not attached)
`DEPOSITION EXHIBIT PAGE
` Exhibit 1005 37
` Exhibit 1009 48
` Exhibit 2001 8
` Exhibit 2006 19
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`5
`
` P R O C E E D I N G S
` AVIEL D. RUBIN, Ph.D.,
` having been duly sworn, testified as follows:
` EXAMINATION BY COUNSEL FOR PETITIONER
`BY MR. OLIVER:
` Q Good morning, Dr. Rubin. How are you?
` A Good morning. I'm fine. How are you?
` Q Fine, thank you. I'm just going to go
`through a little administrative things first.
` Is there any reason you cannot give
`truthful testimony today?
` A No.
` Q You're not taking any medication that
`would affect your ability to give truthful
`testimony?
` A No.
` Q As you know since we've done this before,
`I will endeavor to take a break about every hour
`or so. If at some point you need a break before I
`have called for one, please let me know. All I
`ask is that you answer any pending question before
`we take a break.
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`6
`
` Is that all right?
` A Yes.
` Q And you understand that your attorney may
`object to questions, but absent an assertion of
`privilege you are still required to answer the
`questions posed. Correct?
` A Yes.
` Q I'll try to be clear in my questions.
`That doesn't always happen. If there's any reason
`that you don't understand a question that I've
`asked please let me know, and I'll try and make it
`more clear.
` A Okay.
` Q I'll often be asking questions today
`relative to the perspective of the person of
`ordinary skill in the art. And I believe you're
`familiar with that concept. Correct?
` A Yes.
` Q For simplicity sake, I will likely refer
`to the person of ordinary skill in the art as the
`POSA, just the acronym P-O-S-A.
` Is that all right with you?
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`7
`
` A Yes.
` Q Great. Absent me specifying otherwise,
`please use the POSA's perspective in answering my
`questions.
` How did you prepare for this deposition?
` A I reread my reports and the patent -- the
`patents and the prior art references, and I had a
`couple of meetings with counsel.
` Q About how many meetings with counsel did
`you have?
` A Two.
` Q Two. Do you have any notes with you?
` A No.
` Q Are you emailing, texting, or otherwise
`in contact with anyone else during the course of
`this deposition?
` A No.
` Q I have uploaded and we can provide a link
`to for exhibits, but, you know, if you already
`have certain documents in front of you, that would
`be easier, that's fine with me. So why don't you
`let me know what you have before you before we go
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`8
`
`through the whole downloading process.
` A Sure. I have open my report on the '012
`patent, but I have access to all the others. I
`have Settimi, Evenheim, Landi open. I have the
`'012 and the '651 patents open, but I also have a
`folder with other patents and other reports if I
`need them.
` Q Great. Well, we'll go with the copies
`you have for now. And if we need to transfer
`anything to you, we will.
` Just if you would confirm that the
`versions you're looking at don't have any markings
`or notes or highlighting.
` Is that correct?
` A Yes.
` Q In that case, can you open your
`declaration, Exhibit 2001 for the '651 patent,
`which is for IPR2024-00307.
` A Just a second. Let me find that.
` (Exhibit 2001, previously marked, not
`attached.)
` A Okay, that's open.
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`9
`
` Q And if you go to Paragraph 54 of your
`declaration.
` A Okay.
` Q And if you look at the last sentence of
`Paragraph 4 of your declaration --
` A Oh, I thought you said 54. I'm sorry.
` Q I'm sorry, I did say 54, 54 of your
`declaration.
` A Okay.
` Q The last sentence says, "Although
`tokenized health data has been used for other
`purposes prior to the '651 patent, it has not been
`used in the particular context or manner described
`in the '651 patent."
` Do you see that?
` A Yes.
` Q And what do you mean by "tokenized health
`data has been used before"?
` A The concept of replacing personally
`identifiable information with a token was known.
` Q All right. And what were some of the
`contexts in which that was used or known in the
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`10
`
`past?
` A I'm not sure specifically, but it had to
`do with anonymizing the data, or deidentifying it.
` Q And just so we're clear on the term
`"tokenizing healthcare data," I think you
`mentioned that personally identifying information
`was removed.
` In that usage before, what would a POSA
`typically do to tokenize healthcare data? And
`I'll be more specific.
` What is converted into a token and what
`remains from -- relative to the original record?
` A I'm not sure specifically. If you show
`me a reference, I could discuss it.
` Q Well, I'm just asking you what the
`understanding of tokenizing healthcare data meant?
` A Replacing identifying information with a
`token.
` Q And by "identifying information," that's
`information that would identify the particular
`patient at issue?
` A Right.
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`11
`
` Q So that could be a name or Social
`Security number or address or something similar.
` Correct?
` A Yes.
` Q So what information of the record would
`typically remain after tokenization?
` A I guess it depends on the hypothetical or
`on the specific example that you have in mind. If
`whatever was there that you didn't tokenize I
`suppose would remain.
` Q So if the example was a person's
`healthcare record from a particular doctor's
`office, would what remains after tokenization be
`the specifics of the diagnoses and other details
`relating to the medical treatment received?
` A It depends on what was done, what that
`particular system did to tokenize.
` Q But generally speaking, tokenization of
`healthcare data generally involved removing
`information that would identify the patient.
` Correct?
` A That's my understanding of what could
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`12
`
`have been done with tokenization.
` Q Can you turn to Paragraph 70 of your
`declaration.
` A Okay.
` Q And in the second sentence of Paragraph
`70 you say, "In my analysis here I use the same
`constructions relied upon by the petitioner."
` Do you see that?
` A Yeah, "relied on."
` Q Yeah. So just for clarity, you're not
`offering any alternative constructions for the
`terms listed in Paragraph 70. Correct?
` A Correct.
` Q In that chart of claim terms from
`Paragraph 70, or in Paragraph 70, the fourth one
`down is similarly encrypted.
` Do you see that?
` A Yes.
` Q And what would be a POSA's general
`understanding of encryption?
` A Are you asking me within the context of
`this patent or just out in the world outside of
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`13
`
`this case?
` Q Well, why don't you answer both, and to
`the extent that there's a difference.
` A Yeah. So as we can see from the
`construction, generated such that tokens generated
`from corresponding information can be compared to
`one another to detect a match, so that is -- the
`idea there is that if you have a function that can
`allow you to compare things, then that satisfies
`similarly encrypted. So that could be, for
`example, a hash function.
` In the world outside of this case where
`we're not using the context of any patent,
`encryption has to do with making information
`unreadable to anyone other than someone who has
`the key.
` Q Do you understand the patents at issue
`here to be using the term "encryption" in a manner
`contrary to the regular understanding of the term
`"encryption"?
` A Yes. In the patents in this case there's
`a construction that both sides are using for
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`14
`
`encryption, and it has to do with being able to
`compare things to each other. And also within the
`patents in this case you actually do not include
`the ability to reverse the encryption function;
`whereas outside of this case an encryption
`function would have the ability to be reversed.
` Q Are you aware of anyplace in the patents
`at issue where there's a specific definition of
`encryption given?
` A I'm going to look at the patent.
` Q And for ease of our discussion here
`today, if you could use the '651 patent as the
`example so that we're just being consistent when
`we refer to --
` A Right. That's the one I opened up.
` Q Great.
` A There are several discussions I think
`that led to the definition that's being used in
`this case. These have to do with descriptions of
`the functionality of the encryption server, as
`well as Column 7 at the top where it says, "To
`produce tokens and request tokens, one or more
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`15
`
`encryption techniques may be utilized, for
`example, hash functions and other methodologies
`may be implemented."
` Q Well, sticking with the Column 7 language
`that you just pointed out, that simply refers to
`one or more encryption techniques. Correct?
` A Yes.
` Q And hash functions are just offered as
`one example.
` Is that correct?
` A Yes.
` I should add that the word "encryption"
`in this patent is used in two different ways. One
`of them has to do with the functionality of an
`encryption server, which is to create tokens. And
`the other is, if you look in Column 7, for example
`Line 35, it's talking about encryption for sending
`a request message.
` And in that context it's talking about
`the generally understood type of encryption;
`whereas when it's talking about the functionality
`of the encryption server that encrypts information
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`16
`
`into tokens, that's using the definition that was
`provided in this case. That does not involve
`reversible function; it just has to do with the
`ability to compare.
` Q So I'm sorry, Column 7, what you pointed
`out was Lines 35 through 43. Is that right?
` A Yes. So it says, The request message may
`be encrypted prior to being transmitted over a
`computer network. The request message is
`unencrypted.
` So this is a different type of -- its's
`using the same word "encryption," but it is a
`different function than the one that's described
`earlier where it talks about creating tokens.
` Q And the place earlier where it talks
`about creating tokens is what we just talked about
`at the top of Column 7, Lines 2-5?
` A Well, also, if you look at Column 6, like
`around 54, "Each token obtained from the source
`sites may have been created using the same token
`generator used by the encryption server or using
`any other means that generates the identical token
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`17
`
`for the same personally identifiable information."
` That is a different type of encryption
`than the transmission -- than encrypting for
`transmission.
` Q And -- well, I'll come back to that.
` Is there any place in the patent where
`the term "encryption" is defined in two different
`ways, one for the tokenization and one for
`transmission? For instance, there any place in
`the patent that says, For encryption I mean, and
`then there is a definition given?
` A It's going to take me a moment, because a
`question like is there anywhere in the patent that
`X occurs, I have to look at the whole patent.
` Q Well, let me -- I'll rephrase it.
` Are you aware in your preparation today
`of any place that you already saw that has such a
`specific definition?
` A I think that, you know, when there's a
`construction being used by the parties, then that
`is a definition that's given. And I'm not aware
`of anywhere in the patent that has the words, you
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`18
`
`know, Encryption means X, and defines it that way.
` But I think it's pretty clear from the
`two different uses of encryption that that's what
`is the case in the patent.
` Q Can one create a token using the same
`type of encryption technique that would be used to
`encrypt a message for transmission?
` A Can I please hear that repeated.
` Q Sure. Could one -- could a POSA use the
`same type of encryption that's used for message
`transmission to form a token?
` A No.
` MR. STEPHENS: Objection. Form. Scope.
` Q And are you talking about within the
`context of this patent or just anywhere in the
`field?
` A In this patent.
` Q Talking about generally in the field, if
`one says encryption for data transmission, could a
`similar encryption technique be used to tokenize
`data just in the field of cryptography?
` A I think it depends on whether you're
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`19
`
`operating in a regulated environment like HIPAA.
`HIPAA does not allow for reversible encryption of
`tokens. So if you're asking could you program a
`computer to do it, you could do that. But it
`wouldn't be -- if you were working in a domain
`that required compliance with HIPAA, then no.
` Q With respect to your mention of within
`HIPAA, what specific regulation are you relying
`upon for the idea that only certain types of
`encryption can be used?
` A So I don't have handy, but if you have to
`share with me the HIPAA privacy rule, I could show
`you.
` Q Okay. I'm going to put the -- I'll put
`that document, again just put it in the chat and
`you can download it.
` A Yeah, that's easiest.
` Q So I'm putting in the chat Exhibit 2006.
` (Exhibit 2006, previously marked, not
`attached.)
` A Okay, I've got it. I just need a moment
`to look through it.
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`20
`
` So if you look at Page 2, under C.
` Q Uh-huh.
` A It says that, "A covered entity may
`assign a code or other means of record
`identification to allow information deidentified
`under this section to be reidentified by the
`covered entity provided that," and then 1 it says,
`"The code or other means of record identification
`is not derived from or related to information
`about the individual."
` And so that if you use information about
`the individual to create the token, then this
`would not qualify. And also the last part of 2,
`says that it does not disclose the mechanism for
`reidentification.
` And so if the code is reversible, then
`that would allow for reidentification. So I'm
`relying on the Section C of the privacy rule.
` Q Under C.2 that you just mentioned, The
`covered entity does not use or disclose the code
`or other means of record identification, in a
`reversible encryption system where there is a key
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`21
`
`that can decrypt, would the key that decrypts be
`the code or other means to reidentify the
`information?
` A One more time, please?
` Q Sure. In a public/private key system,
`you can use one key to encrypt information.
` Correct?
` A Right.
` Q And then there's a second key that's kept
`private, in theory, that can be used to decrypt
`that information. Correct?
` A Yes.
` Q And in that system is the private key a
`means of reidentifying or decrypting information?
` A I haven't considered that question. But
`I'm also relying here on the second part, which
`says, And does not disclose the mechanism for
`reidentification.
` And so using that other key would be the
`mechanism for reidentification.
` Q But that second part does not disclose.
`That would only mean if you disclosed the private
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`22
`
`key. Correct?
` What if the private key was never
`disclosed?
` MR. STEPHENS: Objection. Form.
` A So the key itself is not the mechanism;
`the decryption function is the mechanism.
` Q So let me ask this: If a doctor's office
`uses the most up-to-date secure encryption method
`possible to encrypt patient-identifying
`information and releases a record with that
`encrypted patient-identifying information, could
`someone else reidentify the patient without having
`the key, within reason?
` A I think there are techniques used by the
`NSA, such as TEMPEST and others, that could
`interfere with that. But the mere possibility of
`doing a decryption with the key is what makes
`the -- this type of tokenization not HIPAA
`compliant.
` Q Why do you say "the mere possibility"?
` A Well, in Rule 1 it talks about the
`derivation, that the key is derived from the
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`23
`
`information. So that would be one. And the other
`is that there is a capability of reidentification.
`So when it says, you know, it does not disclose
`the mechanism for reidentification, that has to do
`with the possibility of reidentification.
` So my understanding of HIPAA is that if
`you are able to reidentify, to reverse the process
`of tokenization, then it doesn't matter if you do
`it or if you have the key, that is not compliant
`with the standard.
` Q Well, it says, so long as you do not
`disclose the mechanism for reidentification. I
`mean, if no one -- if the person never discloses
`the key for decryption, why would that be
`problematic?
` A Because, one, the key exists and it could
`be found out. Two, the encryption might get
`broken.
` The idea behind this rule is to allow for
`comparisons but not to allow for recovery of the
`information from the token.
` Q So why would you even -- why would one
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`24
`
`even need to have, so long as they do not disclose
`the mechanism for reidentification, if under your
`interpretation the mere fact that it could be
`reidentified is prohibited? That would seem to
`render that language superfluous.
` A I don't see the language you're referring
`to. I don't see "so long as" in the language
`here.
` Q Sorry. It says, C.2, "and does not
`disclose the mechanism for reidentification."
` If the mechanism exists, doesn't that
`assume that the mechanism exists?
` A I'm not sure.
` Q Let me ask another question, then.
` If someone uses the most, you know,
`robust public-private key system that they can and
`they don't disclose the key for decryption to
`anyone, would you agree that the risk would be
`very small that someone could break the decryption
`technique?
` A And just to be clear, we're only
`discussing 2 and not 1 in this example. Right?
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`25
`
`Because you are assuming that the token is derived
`from the user identifying information.
` Q Correct. I'm just talking about 2 now.
` A Okay. So I think the answer is, it
`depends. It depends on the security posture, on
`the random number generation source, on the lack
`of a targeted adversary.
` So, I mean, there are cases where data
`can be protected by parties. But I think the
`reason for these rules is that there are a lot of
`times -- and I've seen it in my consulting
`business -- that someone thinks that they've
`protected information but they actually didn't do
`a great job of it.
` Q But if they did do a great job and they
`did use a good encryption system and they did keep
`the key secret, then they could make it so that
`the risk is very small?
` A I don't think you could say that in a
`general context as that which must be considered
`by someone writing a standard.
` But I agree that there are organizations
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`26
`
`that have more capability in security than others.
`And so if you're asking me is it possible for the
`most well-funded, you know, best-positioned entity
`to do something securely, I think they can do
`better than others. But I don't think anything is
`perfectly secured.
` Q Going back to what you pointed out, C.1
`of the HIPAA privacy rules. The code or other
`means of record identification is not derived from
`or related to information about the individual.
` Let's take the example of -- let's
`actually look at the '651 patent.
` A Okay.
` Q If you look at Column 2 of the '651
`patent, at Lines 44 to 46.
` A Okay.
` Q And that states that the request token
`for each individual may represent encrypted data
`that identifies the corresponding individual.
` Correct?
` A Yes.
` Q So that's indicating that the token
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`27
`
`that's formed is encrypted information about the
`identify of the individual. Correct?
` A Well, it represents it.
` Q Is your understanding the '651 patent
`that the tokenization takes a personal information
`about the individual and encrypts it using some
`method, for instance, a hash function?
` A Yes.
` Q So in that way, the token created in the
`'651 patent is derived from information about the
`individual. Correct?
` A Right. So I see the tension here. And I
`think to answer that, we can look back at Section
`C of the rule.
` Q Okay.
` A And it says that, "A covered entity may
`assign a code or other means of record
`identification to allow information deidentified
`under this section to be reidentified by the
`covered entity provided that," and then 1 and 2.
` So if there is no mechanism for
`reidentifying at all, if that's not possible like
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`28
`
`in the event of a hash function, then this
`wouldn't apply.
` Q That sort of brings us back to 2 then,
`and says, "And does not disclose the mechanism for
`reidentification."
` Why would the rule need to say anything
`about disclosing the mechanism for
`reidentification if under your interpretation it
`can't exist?
` A So when -- so Rule C applies in the
`instance where reidentification is possible. And
`so under 2, it says it doesn't disclose the
`mechanism for reidentification. I'm not sure I
`understand what the problem that you're suggesting
`is.
` Q Well, let me ask this: This Part C
`assumes or provides the manner of allowing for
`reidentification. How would one reidentify the
`individual using a hash function if they had
`encrypted using a hash function?
` A You can't.
` Q What about the doctor's office that
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`29
`
`creates the hash function? Don't they know what
`the hash was, relative to the name they put in?
` A They don't need to recreate the hash
`function because they have the original.
` Q But they would be able to correlate the
`two?
` A They have it all together because they're
`the ones that made the hash.
` Q Just for clarity, a hash function that
`we're discussing about for tokenization would
`take, for instance, some identifying information
`about the individual, such as their name, and
`basically convert that information into the
`resulting token. Correct?
` A That's one way to describe it.
` Q So it's essentially a reconstituted form
`of the input?
` A I'm not sure all cryptographers would
`agree with that, but I see what you mean by that.
` Q And the reason I'm saying that is because
`the hash function is reproducible. Meaning if
`anybody uses the hash function on the same
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`30
`
`person's name, they're going to get the same
`output. Correct?
` A Yes.
` Q So the output has to be based on the
`input?
` A Yes.
` Q Okay. You mentioned that hash functions
`are not reversible. Are there any other
`encryption techniques that are -- other than hash
`functions that are not reversible?
` A There is a class of functions called
`one-way hash function -- I'm sorry, one-way
`functions. And one-way functions, an example of a
`one-way function is a hash function. I'm not
`thinking as I sit here of other examples. But I
`believe in the theoretical cryptography literature
`there are other ways to achieve a one-way function
`besides a hash function.
` Q Then would it be safe to say that,
`generally speaking, when a person says a one-way
`function, they're typically talking about a hash
`function?
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`31
`
` A I think if you're not a professional
`cryptographer, that's probably true.
` Q What about the person of ordinary skill
`in the art in this case?
` A They would likely be referring to a hash
`function.
` Q Can you turn to Paragraph 5 of the '651
`patent, specifically Lines 9 through 12.
` A Do you mean column?
` Q Yes, sorry, Column 5, Lines 9 through 12.
` A All right. Okay.
` Q And that section states, "The tokens 202
`correspond to persons treated by healthcare
`providers but do not reveal the identities of the
`treated individuals."
` Do you see that?
` A Yes.
` Q So essentially one token corresponds to
`one person?
` A Yes.
` Q And if you could turn to your declaration
`for the '651 patent, at Paragraph 56, please.
`
`1 2 3 4 5 6 7 8 9 1
`
`0
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`PLANET DEPOS
`888.433.3767 | WWW.PLANETDEPOS.COM
`
`
`
`Transcript of Aviel D. Rubin, Ph.D.
`Conducted on February 17, 2025
`
`32
`
` A Okay.
` Q And the first full sentence in that
`paragraph at the top of Page 21 states, "The third
`party, for example, using an encryption server,
`generates a request token for each individual in
`the group based on the personally identifiable
`information for each individual in the group."
` Do you see that?
` A Yes.
` Q So that's referring to creating a token
`based on the personally identifiable information
`for each individual. Correct?
` A Yes.
` Q So that might be their name, date of
`birth, and/or other information. Correct?
` A I just have a citation here to the
`patent, so I'm just going to review that.
` I think that's correct.
` Q So in that sense, the token is derived
`from the personally identifia



