#30: Why You Need Patents with Dr. D’vorah Graeser

Guest Introduction (0:00)

Sean Weisbrot:

Welcome back to another episode of the We Live to Build podcast. Today’s episode is so important to every single business owner and startup founder out in the world because patents can literally make or break what you’re doing. That’s why I thought it would be great to invite Dr. D’vorah Graeser to sit down and discuss everything related to them.

Dr. Graeser has a PhD in pharmacology and programmed for the Human Genome Project in the 1990s, an experience there made her realize that she wanted to spend the rest of her life dedicated to helping founders better navigate the global patent landscape. That led her to create KISS platform, which not only provides services as a patent agency, but also provides a suite of AI based tools that assist in competitive intelligence.

Dr. Graeser and I break down everything so you can walk away an expert in patents. We namely discuss what is a patent? Why are there patents in different countries? What is the process of applying for a patent in the US. What’s the cost and time difference between doing it yourself versus working with an agent or patent lawyer? What are the pros and cons of having a patent? At what time should you file the patent? What’s a patent troll? What are the kinds of licenses and their uses? What does the landscape look like now and how will it change in the future? And much more? So, with that, let’s start the show.

Welcome to We Live to Build. My name is Sean Weisbrot and I’m an entrepreneur, investor and advisor based in Asia for over twelve years. Join us every week to fast track your personal growth, so you can meet the ever-increasing demands of the company or companies you are passionately building. Time waits for no one. So, let’s get started now.

Get to know Dr. Graeser (2:10)

Sean Weisbrot:

Really excited to talk with you about patents, something that companies don’t really think about. Before we get started, why don’t you tell everyone a little bit about what it is you’re doing right now, which makes you the right person to talk about this.

Dr. D’vorah Graeser:

Well, thanks so much for having me. I am a US patent agent, which is like being a US patent attorney, except I’m not an attorney at law. I specialize in patents. That is what I do day in, day out, and I’ve been doing it since 1996. So, I have 24 years of experience in this area and I actually got started when I was quite young. I started programming. I got very interested in that, but also love life science. So, after that I got my PhD and I started programming for the Human Genome Project, which was great, except at the time there was a lot of fear because of Craig Venter’s group, Solar Genomics. They were patenting everything left, right and center to do with the human genome.

And many people were afraid that innovators, medical researchers, and others would be blocked from the area. So that’s why I decided to become a US patent Agent, to help innovators outside of companies, or at least outside of large companies, really protect their ideas.

Sean Weisbrot:

Thanks for the intro. So, I’m curious to know what was the end result of that battle? Did he win or did the Genome Project win?

Patent battles (3:21)

Dr. D’vorah Graeser:

Well, in the end, I think a truce was called. Rather than trying to stop folks from getting patents, instead, many of the universities and other institutions got their own patents. And indeed, a number of the patents that Craig Venter got were finally held not to be valid. And this also played out then later on with fights having to do with gene-based diagnostics. You may have heard about some of these fights over the BRCA genes. So, what had happened was there are certain genes that are related to breast cancer and these genes have been patented.

The fear, of course, was that this meant that research couldn’t be done, that new developments couldn’t be made. So ultimately, in the US, it was held that these genes could not be patented, that pure genetic material was not itself patentable, only if it was manipulated by humans, for example, for use in making special proteins could it actually be patentable.

And this is actually in line with what many other countries had long held, which is that pure genetic material could not be considered to be patented. So, you can see there’s always been a balance in the system between wanting to protect the rights of innovators, which I am totally in favor of, but also wanting to be able to help progress and help innovators be able to make new innovations without fear of being completely blocked.

Sean Weisbrot:

It seems like intelligence as an end result, especially because, being from America, I know just how ridiculous patents can be and the court process, and people basically suing each other over basically nothing. So, I’m really glad to know that our DNA is protected if it’s not manipulated.

What is a patent? (4:53)

Sean Weisbrot:

Let’s go back to the main point, which is what is a patent?

Dr. D’vorah Graeser:
Well, a patent is actually a form of property, and that’s one reason why these fights have been so fraught. Patents are the only legally permitted monopoly right outside of, say, highly regulated utilities. Patents enable the patent holder, the patent owner, to block others from making, using, selling, attempting to sell, importing, exporting, et cetera, everything that is protected by that patent, the idea.

Patents protect technology, so they’re a form of property which is meant to protect technology and encourage innovations in the technological area. So, examples of technology might be software, hardware, but also small molecule drugs like normal pharmaceuticals, aspirin proteins like antibodies, methods of producing things like new ways of 3D printing, for example.

And patents were intended to encourage the innovators to share their ideas with the world. They first got started at a time long back in history, when many innovations were held by guild, they’re kept as secrets they wouldn’t be shared with anybody. And then if the guild members died, as was happening, for example, during the plague in the Middle Ages, suddenly those secrets are lost. Quickly, governments realize this is a bad idea, we have to encourage sharing. But if innovators shared, they said, well, we’re afraid. If we shared our ideas, how do we know someone won’t just steal them and make off with them and all of our hard work will go up in smoke. So, patents are always intended to be a balance. They are intended to be a property right to protect innovators, so that they will feel free to disclose their ideas to the world.

The innovator, in return, must disclose their idea freely and they’re given a monopoly, right, for 20 years. So, unlike other forms of intangible or idea related property, let’s say, like trademarks that could be renewed indefinitely, or copyrights that go on for many years after the copyright innovator’s death, patents are only for 20 years.

Types of patents (6:50)

Sean Weisbrot:

What types of patents are there? I’ve heard there’s more than just one.

Dr. D’vorah Graeser:
That’s correct. So, one kind of patent, the most common kind of people think about is the utility patent. That is meant to protect any kind of technology, specifically in the way that it functions. So, for example, it would protect software, hardware, proteins, all the things I mentioned before. Now, design patents protect the way a product looks. So, for example, the way the bicycle looks might be protected by a design patent, but the way it functions would be protected by a utility patent. Both kinds of patents as property can be bought, sold, rented in the form of a license, and you can even use them to get a loan.

There are even special banks set up that do nothing but give loans to high tech companies on the basis of their patents.

Patents in different countries (7:38)

Sean Weisbrot:

I remember when we were having our intro call, you mentioned something about the price of patenting depends on where you patent. So obviously, there’s over 200 countries. They all have different ways in which they govern themselves. But why isn’t there a body that allows you to patent something that’s then protected globally across all the countries?

Dr. D’vorah Graeser:
Well, that’s a very good question. Patents are a per country right. And each country does have its own patent rules. So, patents were set up at the time, basically starting in the 18th century with the rules perfected in the 19th century. So that’s when countries are obviously the main governing unit and they were intended to protect innovators in that country, there quickly arose some treaties to make it easier for innovators in one country to patent in another country. So, for example, innovators all over the world can get a patent in the US. You don’t have to be a US citizen, you don’t have to have a US company, you don’t have to have any US ties. Similarly, anyone who, even if they don’t live in Europe or have a European tie, can get a patent in Europe or for that matter, in China.

But because each country set up its own system and has its own rules, it does increase the expense. It means that you have to file for a patent in every country you want to protection, you have to follow its rules and you have to pay a patent attorney or agent such as myself in that country in order to help you out. There have been moves to try to bring patent rules together, to make them more consistent, to make it easier for innovators in one country to get patents in multiple countries. But unfortunately, there still isn’t one single body that allows you to file in one place and get a patent all over the world. It’s a great idea and I do hope it comes soon.

Sean Weisbrot:

Why haven’t you built it? No, I’m kidding.

Applying for a patent in the US (9:21)

Sean Weisbrot:

Let’s talk from the point of view of someone applying in the US. What does the process look like for starting to finishing a patent application?

Dr. D’vorah Graeser:
So, in the US, as in many other countries, the process begins with of course, having a great idea and a great idea in a technology. So, it has to first of all be in a patentable category. Then, the next thing that you can do is you can do what is known as a patentability search to see if anyone has thought of the idea before. The reason why is, in order to get a patent, not only does your idea have to be in a patentable category, but it also has to be new and inventive. It has to have that wow factor.

Once you’ve done these things, then you go to a patent attorney or patent agent, such as myself, and you work with that person in order to create a patent. Now, what is the patent? Well, it’s basically a document for many ideas, including hardware, software, any mechanical device. You will typically have drawings. The drawings will show the idea, they’ll help to tell the story of the idea. For example, for software, hardware drawings, I recommend that the innovator would have at least seven to ten drawings. It’s a good rule of thumb to make sure you really explain your idea. Then you also have to write the patent text, which you’ll typically do with the patent agent or attorney. And the patent text tells the story of the idea. It describes the drawings, but it also includes a part of the patent that describes the rights that you, as the innovator, as the inventor, want to achieve. And that part of the patent is called the claims. That part is very important because when you go to court, that’s the part that’s going to determine what rights you have.

Once these documents have been prepared, then the patent attorney or agent has to prepare more documents because the US Patent and Trademark Office, the government patent office, where you’ll file, is a bureaucracy. Bureaucracies love their forms. So, after the patent attorney has prepared the forms and you as the innovator sign forms, and you’ve decided that the drawings and the text are right and ready to go, then it’s filed at the USPTO, the US Patent and Trademark Office. And then you wait.

You get a number the same day and a filing date. You have proof. You have a special receipt. So, you know that you filed. But you wait because there’s a backlog. And you wait because you have to wait for an examiner. Humans do the examining. They manually do patent searches to see if your idea is unique, if it has that wow factor, after waiting for quite a while, the examiner will send you a notice saying what they like or don’t like about your patent. And then, you start the negotiation process.

You may have heard the words no sale is ever final. Well, in this case, even if the patent examiner comes back and rejects your idea, you’re allowed to come back with another suggestion. And in the US, you can even have an interview with the examiner, with your patent attorney and your agent as representative. After you’ve gone through this negotiation process. Hopefully at the end, what comes out is you get a patent with claims that specify the rights that the examiner feels that you should have. At that point, the patent is granted and then it issues.
Once the patent issues, you have the right to go to court to defend your idea against others. Now, one point that I want to make in all this, you don’t actually need to have ever built your idea before you get a grant in a patent. You can actually just have a really good design. And it has happened that there are innovators who have come up with great ideas, but who haven’t managed to make a commercial success of their idea and they’ve still been able to defend their ideas in court.

Applying yourself vs with an agent (12:53)

Sean Weisbrot:

Two things you didn’t mention in that statement, well, it was fantastic. But I’m really curious about is what’s the difference in time and cost for doing it yourself versus working with someone who knows the system?

Dr. D’vorah Graeser:
Excellent questions. So first of all, if we’re looking at time, preparing these documents, preparing the text and drawings of the patent can actually be quite tricky. Patents are not written in normal human language; they’re written in lawyer language. They’re written in language that has been honed after years of court cases. So, your patent attorney, your agents will be aware of all these court cases and everything that’s gone in to cause patent language to been shaped in a particular direction. It can be very difficult for a nonpatent person to really get a grasp on that language.

One way around that is to file what is known as a provisional application. You can also do that with the US patent attorney or agent. The advantage of that is the provisional application is easier to prepare and file. It still has kind of confusing forms and also the process of the US Patent and Trademarks itself, just the process for filing is a bit confusing. They do have some tutorials, but it’s a little hard to follow. But it has fewer rules and regulations and you can use it to get a date and then within one year file the final patent application.

Now that final patent application, the one that’s examined, is subject to a lot of very tough rules and it’s very easy to make a small mistake in the process and have the whole thing be rejected. So, the time it can take is, I would say, without a patent agent attorney. I’ve seen folks spend years on it and then come to me to try to fix what they’ve done because they’ve realized that there was a fatal flaw. So, I’m always a bit concerned when folks just do it completely on their own. Doing it with a patent attorney or agent, typically, you’re looking at six weeks to two months. Could be a bit longer if you realize in the middle that you need to rework the idea of your patent or your agent come up with some suggestions and you want to go back and work on a bit more.

So that’s just for preparing and filing the initial application. As far as cost is concerned, it really depends on where you’re doing the filing. We do offer packages from $1,250 to $6,750, for example, for US provisional. If you go to a fancy firm, you can spend up to $20,000 for the initial application depending on how much extra work you do with them. So, the price range is enormous and that’s just for that first filing. And it also depends if you want to go with the provisional. So, it gives you a year to figure out what you want to do or if you want to file that full US application and go forward.

Then the time to actually get the patent, it is going to be faster with US Patent Agent or Attorney because they can do things more easily like arrange an interview or know how to shape the claims so as to get the process to go through more quickly or to see what that initial reaction is and what it means. So, the answer is really that it depends. But I would say if you go totally on your own, you might want to at least go to US Patent Agent or Attorney to review it before filing and to maybe help you out the paperwork because these things can be snares that can block you from even getting the initial filing from being accepted.

How long before your patent gets approved (15:59)

Sean Weisbrot:

And then how long does it take, on average, from when you do the submission to when you hopefully get a response, whether you’re accepted or rejected.

Dr. D’vorah Graeser:

So, once you take the step of filing, not the provisional, but the non-provisional patent application in the US. Typically, now I’m seeing it takes one and a half to two years, or even three or four years before you get the first reaction from the examiner. It depends on what area you’re in. Some areas are severely backlog. The corona virus situation hasn’t helped. We saw a huge backlog where it just seems like everything brought to a halt. Although I say now, they’re coming back on top of the backlog, things are moving faster. So, once you get that first initial reaction, then it depends on you. It depends on what you want to do. You could, for example, decide to quickly call the examiner, have the interview, and just move forward as quickly as possible.

Or you could ask your patent attorney or agent that you’re working with to, “How can we make this process go fast?” Typically, it’s by making the claims, because that’s the part that delineates your property boundaries. Think of it, you want a big fence around 100 acres, or do you want a small fence around like, a quarter acre? Obviously, building a small fence around a quarter acre takes us time. And the same thing is true in the patent world. The more specific protection you want, the less time it will take to get it.

Pros & Cons of getting a patent (17:14)

Sean Weisbrot:

So, with all of the time, energy, and money that goes into filing, working with the agent or the lawyer, and waiting and all of that, what are the pros and cons of even getting a patent?

Dr. D’vorah Graeser:
Well, first of all, being patent pending itself can be quite helpful for a number of reasons. First of all, many investors will view the patent application as being a very important step. Some of them actually insist on it. Others just think of it as being super important. Typically, filing for a patent application even before it’s granted, before it becomes an issue patent, will add a million dollars to the valuation of a startup. And that’s because it represents the potential to block a market niche. It’s viewed as a business tool. And because it’s a piece of property, if something doesn’t go so well for the startup, if it goes under, it can then actually be sold as a way to help recoup the money. And as I mentioned before, you can even get loans on the basis of your patent portfolio.

Now, it also depends on your exits that you plan for both an IPO and for being sold to a larger company. Patents are very important. Facebook actually had to buy more patents before its IPO because the investment bankers handling the IPO said, “You don’t have enough patents. This will make it hard for us to do the IPO.” Because patent pending represents potential, in some ways, it can actually be more beneficial for the startup to actually have a patent application be pending even before it’s examined because then no one can say that you don’t deserve this really broad protection. So, patents help at multiple stages, but in many ways the patent pending helps startups a lot because does it represents potential, as does the startup itself.

Sean Weisbrot:

You had mentioned in our intro call, something about being able to use the patent to protect yourself when licensing it to other companies.

Dr. D’vorah Graeser:

You can buy or sell patents, rent them, which is or rent them, which is like having a license or get a lien on them, which is I like another piece of property. It’s the property analogy.

Sean Weisbrot:

What are the cons of having a patent?

Dr. D’vorah Graeser:
I don’t think there are any cons of having the patent, it does represent the cost. So, before you get, when you do need to think about how it’s going to help you? In my opinion, when patents have not been as useful as they might have been is when they’re only protecting technology, they need to be viewed as a business tool. So, for example, if you wanted to be able to license your idea, that is, to rent it, you would want to think about how you would want to use the patent to do that. Maybe you’d want to use it to work with a larger company as a partner. Maybe you’d want to use it to protect yourself in what is known as channel sales, when you would actually sell your technology through another platform.

Salesforce, for example, has like a big marketplace of apps. So, you could use your patent to protect yourself in that scenario, just to make it clear to the company in question, this is your idea and these are the boundaries and this is what the other company, the channel sale partner or the other partner, what they’re allowed to do with it.

Once you file for a patent application, you also have protection in the sense that if a bigger company, you talk to them and the bigger company then tries to steal your idea, you have proof as of the day of filing that you actually filed for it. I mean, NDA’s, nondisclosure agreements are great, but they’re very hard to defend in court. Whereas a patent is an absolute right and it can be super helpful, especially when dealing with a bigger competitor.

What is the right time to file a patent? (20:34)

Sean Weisbrot:

When we had our intro call, you mentioned the right time to file a patent. It was something about before you start marketing it. Can you expound on that?

Dr. D’vorah Graeser:

Yes, that’s a great point. Thanks for bringing it up. So, in the US, you’re allowed to file for a patent within one year of publishing your idea. So, publication can be anything, could be giving a Ted Talk, could be releasing a product, selling or offering to sell your product, putting up a big description on your website, pitching in public, any of these things can be considered to be a publication.

So, once you make that publication in the US. The one-year clock starts ticking. But outside the US, you are only allowed to file for a patent if you do so before you publish. Now, this is important because businesses become quite international. And actually, I’ve seen this with many of my clients over the years, but it’s really been accelerated in the past year with Coronavirus.

Companies all over the world are looking for solutions, and they don’t care as much if the solution is found in their home country or not. They want the best solution to help solve their problem quickly. So as a result, companies in the US are looking for solutions outside the US and companies outside the US are looking for solutions in the US. For that reason, many more of my clients are thinking about filing internationally. The three big countries in which my clients typically want to file are the US, Europe, and China. I realize Europe is not a country, but because it is one patent office, we consider a country for the purposes of filing.

Patent Trolls (21:59)

Sean Weisbrot:
And this is a term that probably some people have heard of, if not everyone, hopefully. What is a patent troll?

Dr. D’vorah Graeser:

Ah, yes, the patent trolls. Well, a patent troll is an entity which holds on to patents, so they own patents, but they do not practice them. This means that they do not manufacture their own products, they just license them out. This can happen for a variety of reasons. There are some companies that just invest in patents. They buy up patents of failed companies or companies that no longer want the patents to try to put them together in order to make money out of them.

In the case of the BlackBerry, the Research Emotion patent lawsuit, the inventor had died. He had tried to commercialize his ideas and he was unable to do so successfully before he died. And so, his estate then owned the patent, went on to try to get licensing fees to support his family after he died.

In the US, the patent trolls, also known as NPEs, the non-practicing entities, their power has been reduced somewhat because now, in order to get a preliminary injunction, that is when, let’s say, “I accuse you, Sean, of patent infringement,” and I to go to court, but I’m a non-practicing entity, then if I ask for preliminary injunction, if I asked to stop you from doing what you’re doing immediately, then the court would say no. And the reason why is because I’m not in any immediate danger. I am not manufacturing the product. Whereas if I had been manufacturing the product and the court decided there was a very good chance that you, Sean, had been infringing, then I would be more likely to get that preliminary injunction. So that has reduced the powers of the NPEs to some extent, but it is still a concern in the US.

Patent Strategy (23:44)

Sean Weisbrot:

Is there anything I haven’t asked you that you wish I would ask?

Dr. D’vorah Graeser:

Well, a great question, and this is something that you and I had spoken about before, which is patent strategy and failing to take a decision which isn’t itself a decision. Now, what do I mean by that?

At the beginning, when someone has a great idea and they form a startup, you know, they think about their strategy, they think about their go-to-market strategy, they think about their timing. Are they going to raise money or not raise money? If they are going to raise money, when they’ll raise money. All of these things are the subject of deep thought and discussion. Patents should to be subject to the same deep thought and discussion as any other aspects of a business strategy. The reason why is that patents are a business tool. They are a form of property and they are often the most valuable piece of property that a startup will have at the beginning before their product really proves itself in the marketplace.

Because of that, I always recommend that startups, instead of saying “Oh, I’ll think about that later, it’s not important”, and then failing to file for a patent and then realizing two years later that if they had filed for a patent, they’d have a really valuable piece of property, but because they waited so long, they can’t get a patent. Instead of doing that, I always recommend that startups at the beginning, before they publish their ideas, before they go out into that marketplace and as part of their very initial patenting, they think about patent strategy, they think about the benefits the patents will bring to them. They should go out and do their research about what patents are and how they could help them so that they make an informed decision. And if that decision is not to get a patent, at least they’re making their decision with their eyes open.

KISS Platform (25:23)

Sean Weisbrot:

You’re currently working on a platform that’s meant to help people with patents. Does it make the process cheaper, faster, easier? What’s your USP?

Dr. D’vorah Graeser:

So, what our product does, isearch.ai, is it helps innovators understand about the patent and competitive landscape. So, the user puts in their idea and they’re taken on a tour, they’re taken on a tour of competitors with similar ideas and they’re taken on a tour of patents that are similar to their own. And we’re developing new features all the time to make it really clear what the answers mean.

So, in other words, we don’t just throw a pile of results at you, but we actually guide you through the results and explain to you their importance and why you should be concerned about them in order to bring the most value to the innovator.

Sean Weisbrot:

So, I heard the name of your company was KISS Platform. What’s the difference between that and isearch.ai?

Dr. D’vorah Graeser:

KISS Platform is our company, and isearch.ai is the first of a suite of products that we want to develop. So, our goal in life is to help innovators have the same advantages the big companies have. Now, what are the advantages of big companies have? Well, they have resources to do research. They have analysts. The analysts check patents and they say, “These patents are similar. These patents you should be aware of. Ooh, this is a patent pitfall you should not fall into.” And they do the same thing with competitors. What we want to do with KISS platform overall is to help innovators do that same kind of research in a way that doesn’t take a lot of time and that gives deeply guided results.

In particular, with our first product, isearch.ai, we’re focusing on agile, competitive intelligence. So, by agile it is up to date, but it also helps innovators develop their ideas iteratively, meaning, they can go through and as they develop their ideas from a very early stage all the way through to launch. They can come back to the tool again and again, and get more information. Competitive intelligence, because we want to help innovators really find that intelligence that the big companies can find by throwing a lot of resources at it, by throwing a lot of people and a lot of working hours. We want to help them cut through all the noise, find the signal and find the patents and competitors that are most relevant to them.

So, that’s the goal of isearch.ai, but it’s also the goal of this platform. And as we build out more products to do that, we hope to help innovators achieve those same advantages that the big companies have.

How patent systems adapt to change (27:52)

Sean Weisbrot:

So, you mentioned that the whole idea of patenting things started from government several hundred years ago. And obviously, with the Industrial Revolution a lot of things were created, and again, with the internet, things changed again. And now, Industry 4.0 with artificial intelligence and robotics and automation and all these things, the whole world is changing so quickly.

So, for a system that is run by a government and when people assume governments are inefficient, how has the patent system changed? What does it look like today? And what do you think it’ll look like in 20 years?

Dr. D’vorah Graeser:
So that’s a great question. Just one initial aspect that has become super important, it has become more and more difficult for patent examiners to be able to examine patents, particularly in areas like artificial intelligence and blockchain, that are moving so fast.

So, for example, what is an artificial intelligence patent? Our product uses artificial intelligence. Many other products use that. How can patent examiners keep up with it? How can the government patent office keep up with that? And then, how can the courts keep up with that? But then, a second aspect which is really interesting is, how have patents as business tools changed?

One way in which they’ve changed, and that IC will continue to change in the next ten years, will be how folks get together to be able to create things like patent pools. Now, a patent pool is when a group of innovators, typically through an organization, get together and they put their patents into a single pool. And then, everyone who belongs to the pool can use them as a defense in case they are attacked by another company. Or they can also use potentially to defend the products of everyone in that pool if a larger competitor tries to copy them, depending on the rules of the pool.

So, for example, Jack Dorsey, CEO of Twitter, who’s also CEO of Square, has recently founded a new patent pool in blockchain, and he’s encouraging others to put blockchain patents in, in order to keep large companies from monopolizing the space. This is a real danger. The Mastercard CEO recently on a call with analysts, boasted about how many blockchain patents Mastercard has and how they’re going to be so very strong in the space. So clearly, with new technologies, if we don’t want a few big companies to dominate the space, patent pools can be very helpful.

But I think that they can develop further. And one model for this could actually be the telecommunications industry. Now, if you look at your smartphone, it has hundreds and hundreds and hundreds of components developed by hundreds and hundreds of companies, but they don’t all fight each other in court. And that’s because licensing agreements have come up under a regime known as FRAN, which is fair and reasonable licensing, which means that if you want your component to be a standard and to be used by cell phone manufacturers all over the world, you have to agree to licensing terms that everyone else has agreed to, and they keep the costs down and also keep everyone from fighting.

So, in my opinion, in the next ten years, we’ll see more patent pools, more patent agreements come up that will allow for such fair and reasonable licensing agreements. It’ll support small innovators and allow them to get benefit of their work, but it’ll keep everyone from fighting endlessly in court, which quite frankly, is in no one’s interests.

Patents in different industries (30:57)

Sean Weisbrot:

Has there ever been an industry that has collectively decided to not patent anything?

Dr. D’vorah Graeser:

Not, as far as I know, not to collectively decide that. There are industries which are more or less likely to go to court over patents. There are players in industries who are against patents. The EFF, for example, is against the patents, feeling that they stifle software innovation. And there are folks who ascribe, for example, to GNU, the GPL, the general public license, which is again in a strong form, does not allow patent holder to assert patents to those who, if they ascribe to that license, they use software under that license.

But I think that in general, it’s been more of either an all or nothing for individual players. So, some players have said no patents, and other players have said patents all the way. Some companies have decided on their own that they will only enforce their patents if someone attacks them first. So, Twitter has pledged this. Elon Musk in a famous blog post with the title, “All our patents are belong to you,” playing off of the famous meme, “All your bases are belong to us.” He said that, “Look, Tesla doesn’t want to stifle innovation, but it doesn’t want to get attacked. So, as long as you don’t attack Tesla with your patents, it’s not going to attack preemptively with its own patents.” It wants to come to some kind of sharing agreement, and that’s fine for individual companies, but it still leaves a very big gray area.

What does it mean to share patents? What does it mean to compete? If a small startup says, I don’t want to enforce my patents, but it’s bought by a bigger company, what does that mean? So that’s why I think more defined agreements would be useful. If an entire industry decided that it didn’t want patents and it put that into an agreement, at least it would be clear. But the situation right now is super unclear. There isn’t any kind of yes-no. There are a lot of folks attempting to find new solutions, but I haven’t really seen anything take hold that could be a model that every industry could ascribe to.

Sean Weisbrot:

I know from human psychology that it’s going to be hard to get any industry to be so benevolent.

AI & patents (33:05)

Sean Weisbrot:

With that in mind, what is the artificial intelligence patent landscape look like? Is it possible that there’s going to be these agreed upon standards? Or do you think everyone’s going to make their own models and patent them? Or do you think we’re looking at an open-source style system? How do you think AI will play out?

Dr. D’vorah Graeser:

That’s a really interesting question. AI is a field that brings with it its own difficulties. So one is, can an AI be an inventor? The answer is probably not. But you may have seen these rather lovely images of car chassis that were designed by an AI. So, the chassis is like a cage designed to protect the driver. Nowadays, even though cars are very light, driver fatalities have gone way down because the chassis is built as a cage, a very protective cage.

There was a recent example of this with a race car driver. His car flipped over multiple times at extremely high speeds, burst into flames. He walked out of it unscathed. And a big part of the reason for that was the cage around him was built to be so protective. So, they asked an AI to design such a cage and it looked like bone. I mean, it was gorgeous. It was almost biological organic looking. It looked radically different from anything a human had ever designed.

So, the question is, is the AI inventor? The answer is probably not, but no one’s really certain. And quite frankly, I think the courts are kind of hoping this question doesn’t come up, because it’s going to be such a hard one to answer. So that’s one big question that we’re all looking at, but no one really has an answer to.

Another big problem is many AI models are black box, meaning it’s very hard to understand what’s going on inside them. Famous example of that is the software that was designed to find coyotes in the snow from all these pictures that were taken from a motion sensing camera. Turned out that what the AI thought was a coyote was black nose, tips of black ears, black tails, and snow. Because all the pictures of the coyotes came with snow. So that just shows you the pitfalls of AI, but also the pitfalls of examining AI models.

Now, it could very well be that folks will reach agreements in particular industries. I have a hard time seeing that the entire AI industry and all these different verticals will come to a single agreement. However, I do hope we have discussions when we’re discussing things like AI ethics. We also talk about AI patents and how should AI models be protected and how much information should really be disclosed and what should the inventors know about their idea before they try to patent it. But so far, no clear answers to any of these questions, Sean.

Sean Weisbrot:

So, do you think in order to be able to claim that an AI can be an inventor and therefore on a patent, that first we’d have to come to the conclusion that AI is sentient?

Dr. D’vorah Graeser:

Oh, I don’t know about that. I think that may be above my pay grade. However, I would say that we would have to consider who an inventor is. Right now, an inventor is considered to be a person. You may have remembered the monkey selfie controversy where a baboon was allowed to play with photographer’s camera, took a picture of itself with the camera. And this became known as the monkey selfie and the photographer went to court against a wildlife organization that said, “No, we’re going to represent the monkey. And the monkey is actually the owner of the copyright, not the photographer.” And the judge says, “No one owns a copyright. A monkey is not a human. Only humans can be creators,” which would seem to mean they can only be inventors. And so, the photographer doesn’t own the copyright and the monkey doesn’t know the copyright. Now, in the case of AI, what would that mean? Hard to know.

Patent licensing (36:28)

Sean Weisbrot:

That sounds speciesist. I don’t like that. I think the baboon should have its own protections. So, there was one more thing that came to mind when we were talking about all this, which is licensing. So, you mentioned several times that it’s possible to license your patent. What kind of licenses are there?

Dr. D’vorah Graeser:

That’s a great question. They actually represent a contract. So, it’s as though you’re going to go rent an apartment. There are more or less standardized apartment contracts wandering around, but they depend on the laws of your state. In the case of patent license, they would typically be governed by federal law, but in the US state laws can also apply. So, you’d have to think about that. You’ll go to an attorney in your state of choice to develop that license and also to understand how it applies at the federal level.

When you’re thinking about a patent license, it’s really an agreement where you’re agreeing to rent out your idea to allow someone else to use it. Maybe also with know-how, maybe with code, for example, for software, maybe with hardware or with sales of hardware. It can include many different components, but it is part of an overall business agreement.

So before starting to even work on license agreement, I would suggest that the company thinks about what its goals are for the business agreement. That would include that licensing agreement. Does it include sales? Are you selling any kind of free to use of the idea, or are you selling use of the idea only in a very specific way? Once you decide what your business goals are, then developing the license becomes much easier.

Important life lessons (37:59)

Sean Weisbrot:

What is something that you’ve learned recently and how are you putting this into practice, either personally or professionally?

Dr. D’vorah Graeser:

One thing that I have learned recently, which is really important is the need to have a plan, but be willing to change your plan quite flexibly and of course, with the whole coronavirus. So, we started the year thinking we’re going to develop our AI slowly, which is of course, KISS Platform, our tool to help innovators understand the patent and competitive landscape. We thought we’d develop it slowly; we’d release some modules.

When the coronavirus hit, we started getting a lot of calls from clients and others saying, we need help. We need to know more about patents. We need to know more about our competitors. And people were shut off from their usual avenues from being able to do that. Like, they couldn’t go to meetups, they couldn’t go to conferences. We were in the same boat. So, we quickly changed our plans while listening to our users. It didn’t mean we allowed ourselves to go into kind of a free-for-all or, we simply stood there like deer frozen in the headlights.

But instead by listening to our users and listening to others and realizing what we could actually do now, no, we couldn’t go to in person conferences. That was just off the table. We couldn’t go to in person meetings so easily. That was just off the table. Dealing with our current reality, the situation we were in, and thinking about how we could change our plans and go forward while always listening to our users. We’ve tried to do that all the time. Because the situation on the ground right now is so fluid, Sean, that we’ve had to learn how to be very flexible and to be comfortable with this kind of continuous change and uncertainty.

Sean Weisbrot:

And I think that’s the theme of the 2020s.

Dr. D’vorah Graeser:

I totally agree. I am hoping that 2021 is going to be better for us all, but of course it will introduce new uncertainties as we hopefully transition to the post-covid era.

Follow up with Dr. Graeser (39:46)

Sean Weisbrot:

So how can everyone follow up with you?

Dr. D’vorah Graeser:

First of all, you can find me as D’vorah Graeseror KISS Platform on LinkedIn. We are also @KISSPlatform on Twitter. You can also find me at info@kissplatform.com. That’s Kiss, K-I-S-S platform all in word, dot com. And you can also find us at isearch.ai, that is our product, or kissplatform.com, our main website.

Sean Weisbrot:

We’ll have everything on the show notes at welivetobuild.com/listen. If you liked this episode, please make sure to reach out to her, as well as leave us a review on iTunes or Spotify or wherever it is you listen to this podcast. We appreciate your time and your energy. Entrepreneurship is a marathon, not a sprint. So, take care of yourself every day.

Thank you, Dr. D’vorah Graeser for your time.

Dr. D’vorah Graeser:

Thanks so much for having me, Sean. It was a lot of fun.