The IFI Interview: Patents on the Blockchain with IPwe’s Jonas Block

It's hard to remember a time when Bitcoin, which today is tracked on market tickers right next to the Dow, Nasdaq, oil and gold, was not part of the humdrum mainstream financial consciousness. But the 2017 crypto craze that gripped the market with Bitcoin's dramatic rise from some $1,000 a coin to nearly $20,000 during the year, intrigued IP attorney and investor Erich Spangenberg, founder of IPwe, a global financial technology company that leverages cutting-edge technology to enable businesses to understand their portfolios from a financial viewpoint.

Unlike other impassioned investors, Spangenberg understood the 2017 Bitcoin bandwagon was mostly hype. But he also understood something deeper, though less dazzling: the blockchain technology underpinning Bitcoin makes transactions faster and easier to execute and has the power to transform the way many companies do business. Applying that thinking to IP, he saw that recording patents on the blockchain could make the excessively complicated industry—with its myriad of country and regional patent offices around the world delivering data, each in its own way—more transparent. Transparency naturally leads to more liquidity in the market.  He founded IPwe in 2018, with the aim of harnessing blockchain technology and artificial intelligence to address illiquidity difficulties that come with the lack of data transparency in the IP space. And the company's blockchain machinery is forging new paths. In 2021, it released an NFT (non-fungible token) marketplace for patents—yes, the same blockchain-supported certificates that speculative collectors had been wildly chasing in the digital art world over the past two years.

IFI CLAIMS has been riveted by IPwe's leaps and bounds in the arena and wanted to learn more about the company's technology, so we reached out to Jonas Block, who heads product management for a fascinating chat about how school sparked his own patent journey, why IPwe's patent NFTs enable liquidity, and how the company once made an attempt to switch data providers before understanding how much they need IFI's clean, trustworthy patent data. Spoiler alert for any of you considering something similar: it turned into a “huge mess,” according to Block. He doesn't recommend it; neither do we. The interview below is edited for clarity and length.

IFI CLAIMS: What is your background, and what led you to the patent industry?

Block: I got into the IP industry because I went to school with children of patent examiners, so I was very interested in patents from the cradle, so to speak. Later in my education, I studied law here in Germany, and became an IP litigator for almost eight years, before earning my Ph.D. I wrote my thesis on automated SEP licensing as well.

Then I joined IPwe, originally as head of AI development. I was tasked with how to structure patent data in a way that an AI engine can read it and return high-quality results. I was in that role for a year and a half before I took over product management at IPwe. I work 100% in patents, even though IP is much bigger than patents.

IFI CLAIMS: We have to ask: how did attending school with children of patent examiners get you interested in patents?

Block: It's actually a funny story because when I was a kid, the patent examiner kids used to bring false prints, or misprints, from the patent office to school to save on paper. And instead of doing math, I read the flip side of the papers and actually became intrigued by this weird language that is used in patents. To me, it was like a riddle, and I wanted to solve it. Later, I had some IP overlaps with music, and figured out that IP is a much greater playground. Copyrights and patents and trademarks aren't tangible, but they're super relevant in our daily lives.

IFI CLAIMS: Why was IPwe started?

Block: The IPwe history officially started in 2018, but the technology is much older. I'll give you the technology prequel. Erich Spangenberg, the founder of IPwe, has been working in the patent industry since 2003. In 2007, he purchased a predictive analytics company from the University of Minnesota. At first, this technology was focused on freight containers–predicting which ones should be supervised by customs.

From 2008 onwards, Erich used the analytics tool to build out a patent analytics engine. Over the years, he developed and refined the algorithm to better understand patents and to predict their value. That engine became IPwe in 2018.

Erich was also intrigued by the Bitcoin high in 2017. And though he saw that this was primarily hype, the underlying blockchain technology could be put to an actual use if patents were recorded on the blockchain. Originally, he had the idea to build a global patent registry, because IP registers are national and there's so little transparency. He looks at patents from a financial angle and thought patents should transact a lot more than they actually do. One of the reasons why they don't? The huge cost overhead of understanding who owns what and what is active in which part of the world with this grid of national patent registers. So he was trying to see at the global level how a blockchain company can quickly facilitate access to knowledge locked up in patents. That way, you can identify what might be valuable. We've been on that mission since day one.

IFI CLAIMS: Tell us about your technological process of turning patents into NFTs. Why do you need to do that?

Block: I was intrigued by the whole concept, which is why I applied for a job at IPwe. The whole concept of an NFT is relatively new, so it's not like we knew right from the start how to do it. We started out just creating blockchain records for each patent, but that's nothing more than an Excel spreadsheet basically, so you can't really do anything with it. And you can't transact the patent. So the idea was, how can we find a shell that houses the patent? And if we want to transact the patent, how can we make sure that the shell can be moved from owner A to owner B?

There are benefits around blockchain technology. It's auditable. It's immutable. But you have to set it up in the right way to yield those benefits. The NFT is an appropriate solution for this because you have this combination of a token that represents the title to a specific IP asset. And then you have the metadata document that holds all of the relevant data points. We didn't just call it an NFT for the sake of surfing the NFT wave. It was actually what we had been looking for—for a while. We were very skeptical in the early days because of NFTs like Bored Apes, for example. You can't just attach a picture stored somewhere on the web and then say, I own this now, because you can't really do anything about it. At the time, there were a couple of people on Twitter saying: Look, I copied your picture; what are you going to do about it? Remember, there were no IP rights assignments to these things. So we understood the potential that was in NFTs, but we wanted to make sure that we converted the patents into something that was using this technology the right way. We spent a lot of time thinking about the metadata document which is the link to NFTs. That's the heart of our digital asset.

IFI CLAIMS: How do you protect your NFTs from being copied?

Block: The problem with copying doesn't sit on the NFT itself. That lies with the real-world asset that is connected to the blockchain representation of the asset. In our case, it's a patent.

Outside of the U.S, for digital art, you don't have to register for a copyright, which means there is no real-world asset connected to this copyright. You just frame it at the point in time when you create a token. And then you have to assign rights. From a legal perspective, if I own a house, but I'm not allowed to throw anybody else out of my house that I own, then this property is worthless to me because anybody can walk into my house. That was the problem we had with NFTs—there was no rights assignment in the sense that I can throw people out for the digital art NFT. So this is why they were basically like collectibles.

For patents, we had the patent register in mind right from the start. So we had to pay attention because if somebody tried to copy the NFT, that's not possible because it's unique in the sense that it has a unique blockchain hash. What you could do though is you could try to create a second NFT for a patent or you could try to reassign a patent on the real-world register without reassigning the NFT so that the two rights fall apart. Something we have been paying a lot of attention to is tracking the patent registers in order to consistently synchronize them to the NFT ownership. Ultimately, it would make sense if the whole IP system embraced the concept of NFTs representing patents.

IFI CLAIMS: Why does that make sense?

Block: Let's assume a patent gets sold from Company A to Company B. If everything was blockchain compatible, that assignment of the NFT from Company A to Company B would be updated in the patent register as well. So it would be impossible for the two to fall apart. Right now, what can happen is that you have Company A owning the NFT and Company B being registered in the patent register, so in this case, the NFT ownership needs to be suspended, or we need to flag that this is not in sync. This is something that we're using IFI data for a lot because we're constantly monitoring whether the current owners of our tokenized digital assets, which are patents as I said, are also the registered owners. This minimizes the risk of the two legal positions falling apart.

IFI CLAIMS: Let's zoom out from the details of tokenizing patents. What general problem is IPwe solving for the market?

Block: The problem we're addressing is that IP data is extremely siloed. You have bibliographic data that is available in the patent registers and IFI does a great job at cleaning and standardizing this data, for example. But you also have other data providers. Litigation data is separate. You have royalty data, prosecution data, which is a different data set. You have pre-filing data, you have data that is distributed in law firms, where litigators are dealing with it in court registers. Financial data sits with banks for security assignments. This data is scattered all over the place. Tokenization of patent assets puts all of the data into one single source of truth, which is the metadata document of our NFTs that holds all of the data trail. This minimizes discovery costs.

When I was a litigator, a client would come to me and say, Hey, I want to sue a company based on patent one, two and three. I would go to the patent register, look up patents one, two, and three. I'd go to the USPTO and look up the file wrapper and so on. While I'm collecting data, I'm wasting legal time, which is extremely expensive. With our platform, in the due diligence process, you don't have to get that data all the time. You can simply point your attorney to the NFT and tell the attorney that all that you need to know is in this little warehouse on the blockchain, and I'm going to grant you access. The attorney can consume it, and doesn't have to spend tons of time searching for this data.

Another advantage is verification of data points, or going back to the source of truth. What we're doing is putting verification on each of the data points to identify ones you need to look at more closely to separate which data points might be flawed and which are not. The result is a reduction of the transaction costs that can be suffocating to small and medium-sized entities.

IFI CLAIMS: Where does IPwe fit in the patent landscape? Is tokenization and blockchain the reason why your customers come to you?

Block: If we look at the broad IP space, there are a number of companies doing tokenization of copyrights. There are a couple of companies, in addition to European authorities, doing tokenization of trademarks and industrial designs and putting them on the blockchain. There are a couple of companies going into the patent space with tokenization or going into the pre-filing space, which means they're tokenizing ideas and storing those ideas on the blockchain.

The competitive edge of IPwe—our product market fit—is not simply that we've been doing this longer than others in terms of collecting, cleaning, verifying, and then storing the data in our systems. We are not a mere data provider that allows customers to take it and do something with it. We have built algorithms on top of that data that do valuation or validity analysis of patents. If you think of this like a pyramid, we have this data foundation, which is extremely important, but that can be recreated by other companies if they really wanted to pull in bibliographic data.

So, it's standardization of the data. Then verification of the data. And then we are digesting this data in the next layer, which is the AI engine running on verified, clean data. And that's our competitive advantage because it is not easy to recreate. For one, we've put a lot of funding into our AI engine. Secondly, the way we're returning the data to the customer is unique—the AI actually shows you what you need to pay attention to. We've designed great transparency into the system because we're targeting a product for the financial industry right now. Financial professionals are looking for trustworthiness of information; they need confidence about the specific valuation that they're getting. IPwe derives this from the level of integrity and verification of the underlying data. It's interoperable. And it's transparent to the user. Our results are the opposite of ChatGPT, where you type something in and you don't really know what it's doing in the background. We summarize the data and express it to the customer in a way that makes them confident. It's not a black box.

IFI CLAIMS: What's the value proposition to your customers to tokenize this patent data?

Block: It's 100% about transaction costs. If you want to sign a license agreement, you have to send contracts around. You have to send claim charts around. You have to discuss the license with a specific customer and ultimately sign a license agreement. Your legal department, or IP department, administers that license and tracks it. And when they think it's no longer required, they terminate it, and they file it away somewhere.

If you have a tokenized asset, you are now in a new environment of smart contracts or self-executing contracts. Contractual requirements can be recorded on the blockchain; no human has to touch it. If you record on the blockchain that the license is active when the money hits the bank account, then you don't have to have a human reviewing this anymore. Basically, the bank account says “money received,” sends it over to the smart contract, which reads the blockchain, and then the blockchain says, “license active.” In that way you can have microtransactions enabled.

For big corporations like Qualcomm, for example, that have a huge, established licensing program, it's less interesting because they already have established processes in that space and they have a well-oiled, due diligence machine. But where it is very interesting is for small companies and supply chains that currently don't have access to it because the costs are just prohibitive.

IFI CLAIMS: IPwe talks about “abstracting value.” What does that mean?

Block: Value locked up in patents is extremely subjective. A bottle of water in a desert has a different value than a bottle of water next to Niagara Falls. With patents, the value can depend somewhat on what the customer wants to hear. What I have experienced is that if the customer wants the valuation to be a dollar, it's possible to show data that the patent valuation is a dollar. If the customer wants to have a valuation of $1 million, it is possible to approximate this if you tell the right story around the data points, and if you don't have a major flaw that limits your value.

This is problematic because during a transaction, Company A and Company B can be very far away from each other because one wants to push the valuation price down, while the other wants to push it up. There's no baseline valuation without collecting tons of data points and then getting into the storytelling. The mission that IPwe pursues is trying to abstract value locked up in patents. There will be people saying there is no such thing as an objective value. But what we're doing is mass analysing patents. Let's say you're looking at the patents in drone technology in China. You take a look at one patent relating to drone technology, and you're doing evaluation for that one target that you want to understand. But then you need to do a patent landscape around this to understand what else exists in the drone patent space in and outside of China.

Let's say you find 1,000 other patents related to drones in China. You can't look at those 1,000 patents because it's too much. So, you need to find a way to create a snapshot of those 1,000 patents and look at all of them because they have a relative value. If I'm the only patent owner in a specific space, that's valuable. But if there are 40 others similar, my patent is of less value.

IPwe is taking an objective approach to every single patent, independent of where it's from and who owns it. We're looking at which market the patent plays in and the relative strength of the patent against peers. This way, you can be sure that we have looked at all the patents in the same level of detail. We create the baseline value of the patent. This is very helpful for financial orientation in the space. And nobody does that.

IFI CLAIMS:  Are there any general trends you're seeing in patents globally?

Block: I think there are a couple of trends that are relevant for everybody in the IP industry. Quality over quantity is going to be the topic of this decade. As AI catches up, it will be easier to see through low-quality patents, so it won't make sense to maintain them anymore because you're basically wasting your money. It won't be a shield to your business anymore. As a result, companies will focus on filing more high-value patents going forward and filing numbers will decrease. IBM announced that they're not trying to become the world's biggest patent filer anymore because they care more about quality.

I think we will also see a diversification in the territories where patents are filed. For example, in the recent litigation between Ericsson and Apple, you had a preliminary injunction from Colombia because Ericsson maintained patents in Colombia. That was a game changer. Companies understand that they need to file broader so as not to give away too much knowledge in the public domain.

The third trend is that trade secrets are increasingly on the rise. At IPwe, we file for patents for our AI engine all the time, and in the interest of transparency, we disclose which data points we use for what, but instead of publishing blueprints, we keep the code base secret and the process of how we're doing it. We are not open sourcing it—it's a trade secret.

The fourth trend that I think is going to be relevant is the integration of technology areas. In the past, you had one patent for your toaster, and you had a second patent for your car headlight and you had a third patent for your screwdriver. Now, everything is smart, so you have patent areas overlapping. Suddenly your fridge infringes telecommunication patents. Suddenly you have electronic components in your screwdriver. Everything is merging, which is why you have a large licensing potential in many spaces.

Integrate key financial indicators with IFI's global patent data

Sound investment decisions require you to go broad and deep. Analyze your investments in the most strategic way you can with CLAIMS Direct. We have the updated and accurate patent data you need to understand how your target is moving in the market. Patent-based analysis leads to better capital allocations.

Learn More

IFI CLAIMS:  Why do you use IFI data?

Block: We like the quality. We once made an attempt to discontinue working with IFI data. We tried to switch to a competitor, but it was a huge mess. Sometimes, when you're on the outside, you can underestimate how much work IFI puts into aggregating, cleaning, and standardizing the data. There are EPO, USPTO and WIPO portals with data access points. And we thought at the time that it would be easier to build a system with the data exactly how we needed it so that we could have fewer algorithms sitting on top of the IFI data. We paid to develop this idea, and it did not work out.  And we gained a completely different level of appreciation for the IFI data. We tried to leave, but we came back because we understood that the IFI data set has tremendous value, and that starts with the quality of the data.

We've noticed this in the new IFI machine translations for Asian documents. They have increased in quality because the technical people on translations are extremely good. IPwe's AI works in English, so we're relying on these translations.  This is where a lot of the advantage is on IFI's side. These are trustworthy. We have checked them.

The country Snapshots are tremendously valuable. The territorial coverage is very good from IFI. We learned that to orchestrate all of this is extremely difficult. For us, submitting to the server and the AWS integration is just very easy to plug into and get going. That's why we are with IFI.