Value Driven Data Science: Boost your impact. Earn what you’re worth. Rewrite your career algorithm.

00:00:00 Dr Genevieve Hayes
Hello and welcome to value driven data science brought to you by Genevieve Hayes Consulting. I'm doctor Genevieve Hayes. And today I'm joined by Helen McFadden to discuss the intersection of artificial intelligence and intellectual property. Helen is a Patent and Trademark attorney with a background.
00:00:20 Dr Genevieve Hayes
And artificial intelligence and mechatronics engineering, she has successfully obtained patents, trademarks and designs for businesses in Australia and overseas in a large number of technology areas including machine learning and image classification automation.
00:00:38 Dr Genevieve Hayes
Smart devices audio signal processing, embedded software and control systems. Helen, welcome to the show.
00:00:46 Helen McFadzean
Hi, thank you for having me.
00:00:49 Dr Genevieve Hayes
If you look at the list of the greatest inventions of the 20th century, you'll find they all have one thing in common, from tea bags to toasters and from cellphones to cellophane.
00:01:00 Dr Genevieve Hayes
They all take the form of physical objects and are, or at least were protected by patents. Yet since the turn of the century, the nature of inventions has changed significantly, and many of the greatest inventions of the 21st century take the form of computer code. But how do you protect an invention?
00:01:21 Dr Genevieve Hayes
When you can't physically touch.
00:01:22 Dr Genevieve Hayes
It that's something we're going to be discussing in today's episode, but before we begin.
00:01:28 Dr Genevieve Hayes
Whenever I think of lawyers, the first thing that springs to my mind at trial attorneys, I just rewatched a few good men last night.
00:01:37 Dr Genevieve Hayes
So you got Tom Cruise as a trial attorney there and you have all the TV legal dramas, you know, many years ago I wanted to grow up to.
00:01:46 Dr Genevieve Hayes
Be Ally mcbeal.
00:01:48 Dr Genevieve Hayes
And you've got shows like Bull and Boston Legal, but intellectual property law like it seems to be something quite different.
00:01:57 Dr Genevieve Hayes
In fact, one of the things that I was really fascinated about that you told me when we first met was that the basic qualification to become a patent lawyer is actually a technical degree, such as engineering rather than a law degree.
00:02:13 Helen McFadzean
Yes, that's right. So partner attorneys.
00:02:16 Helen McFadzean
Are quite different to lawyers, and I suspect lawyers are quite different to lawyers on TV.
00:02:21 Helen McFadzean
So patones are a very specialised profession in order to become a patent attorney, you need to first have a degree in either science and engineering, and then you go on to do very specific specialised studying in intellectual property law and you also need.
00:02:42 Helen McFadzean
Two years of work experience at a IP firm and fill out a thing called a statement of skill. Before you can become registered.
00:02:49 Helen McFadzean
So it's it's a very, very specialised area. I guess if you if you think of intellectual property law as a as a specific area of law in general patent attorneys and patent law sits in an even more specialised.
00:03:01 Helen McFadzean
Area of intellectual property law. And then if you look at patent attorneys as a group, you know, everybody's got different science and engineering backgrounds.
00:03:09 Helen McFadzean
So you've got patent attorneys that specialise in, say in my area, software, electronics, ICT type inventions, you've got partner attorneys that do chemistry and life sciences and you know separately mechanical.
00:03:21 Helen McFadzean
Engineering as well. So we're specialised and then specialised super more in our specific technical areas in our specific area floor.
00:03:30 Dr Genevieve Hayes
So you initially did a degree in Mechatronics engineering. What made you decide to go into IP law rather than pursuing a career as an mechatronics engineer?
00:03:41 Helen McFadzean
That's a really good question. So this is gonna make me sound really old, but back in my day when I finished my mechatronics engineering degree, gosh, it would have been maybe 2008, I think 2007 or 2008. I did my engineering degree in Auckland.
00:04:00 Helen McFadzean
University and when I graduated, a lot of my peers either just went in.
00:04:05 Helen McFadzean
To graduate roles for either mechanical or electrical engineers, the industry you know, in that control systems and you know, mechatronics, engineering, it was wasn't there were not that many sexy jobs like there are today 4 graduates from mechatronics engineers. So I actually went into.
00:04:26 Helen McFadzean
This advisory I got a job with Ernst and Young for my first year out of university and I felt that I wasn't really making use of my technical background and I still remember. So one of my good friends out of engineering, he.
00:04:43 Helen McFadzean
Went to work for New Zealand IP firm called Asia Park and I remember catching up for lunch with him was and I was just like Oh my God, I don't know what to do with my life.
00:04:51 Helen McFadzean
You know just what.
00:04:52 Helen McFadzean
Do I do I go back?
00:04:53 Helen McFadzean
To engineering, you know I'm not. This is not really the right career choice for me.
00:04:58 Helen McFadzean
I'm doing business advisory and he said, you know, why don't you try IP and I you know it's kind of.
00:05:03 Helen McFadzean
Least you get to use your technical background. He he went through with me. What? You know you would need to do to become registered.
00:05:09 Helen McFadzean
It's a bit of engineering. It's a bit of IP, which is something new. There's a bit of business involved.
00:05:14 Helen McFadzean
You kind of need to work with clients and trying to understand their commercial objectives and how you can use intellectual property to help them achieve that.
00:05:21 Helen McFadzean
And it sounded really interesting to me. So I looked into it more and Fast forward 14 years. I'm I'm still in.
00:05:29 Helen McFadzean
Intellectual property so.
00:05:31 Helen McFadzean
I thank him for that good advice.
00:05:33 Dr Genevieve Hayes
I personally find the whole intellectual property law discipline fascinating. A few months back, I was visiting my parents and I ended up with this huge conversation with my dad about how Disney uses IP law to protect its characters. I don't suppose you've heard of the movie Winnie the Pooh blood and honey?
00:05:54 Helen McFadzean
Not the movie, but I've read a few Winnie the Pooh books to my 3:00.
00:05:58 Dr Genevieve Hayes
Year old the back. Your three year old this.
00:05:59 Dr Genevieve Hayes
Not this is not.
00:06:03 Dr Genevieve Hayes
Is a Winnie the Pooh horror movie. Ohh.
00:06:07 Helen McFadzean
No, I have never. I didn't realise such a thing existed. I'm going to have to go look it up now.
00:06:15 Dr Genevieve Hayes
Anyway, my dad and I were in JB Hi-fi and we came across this copy of Winnie the Pooh blood and.
00:06:20 Dr Genevieve Hayes
Honey and our immediate thought was how have Disney lawyers not shut this movie down? Ohh, that's an interesting that's an interesting question. Well, so it's it's that's more copyright.
00:06:34 Helen McFadzean
Which is not my area of expertise. I do know like a tiny tiny bit. What I suspect has happened is that because there's some some exclusions and copyright infringement. So you know, parity and satire those type of things may be consider.
00:06:49 Helen McFadzean
But either I can't. I don't. I can't remember the specific wording for the provision. Something along the lines of Fair use or fair dealing equivalent to those types of provisions that you know, there's certain things that you can do that exempt you from copyright infringement, maybe that in this particular case because it's parody and satire, it falls within that exclusion.
00:07:10 Dr Genevieve Hayes
We actually did.
00:07:11 Dr Genevieve Hayes
Google this and we found out the answer to this one. We need a piece now out of copyright.
00:07:16 Helen McFadzean
Ohh OK well that that that explains it too, because it's been 70 years beyond the death of the author original.
00:07:25 Dr Genevieve Hayes
I think it's publication date or whatever. Anyway, the point is Winnie the Pooh. The first book is out of copyright.
00:07:32 Dr Genevieve Hayes
The second book isn't so you can make a horror movie with all the characters.
00:07:35 Dr Genevieve Hayes
From the first book.
00:07:36 Dr Genevieve Hayes
But you can't include Tigger because Tigger didn't appear until the 2nd book.
00:07:40 Helen McFadzean
Right.
00:07:42 Dr Genevieve Hayes
I'm still amazed that Disney didn't find a way of shutting this down, but I'm pretty sure if there was a way.
00:07:49 Helen McFadzean
But maybe it's, you know, it's it's promoting Winnie the Pooh inadvertently for them. So you know, it could be a win win for both.
00:07:56 Helen McFadzean
Parties, who knows?
00:07:57 Helen McFadzean
Not sure, not sure depending on the movie. I haven't seen it.
00:08:00 Helen McFadzean
So I don't know.
00:08:02 Dr Genevieve Hayes
I haven't actually watched it yet, but I do have the copy because, as I said, my dad and I were convinced that Disney would be working really hard in order to destroy every copy of this.
00:08:12 Dr Genevieve Hayes
So we figured buy it immediately before it's no longer.
00:08:14 Helen McFadzean
Yes, because it could be a collectors item, yeah.
00:08:17 Dr Genevieve Hayes
But you know, this raises the question. You've got your patents, your trademarks, and your copyrights, and I understand that copyright does apply to things like Winnie the Pooh.
00:08:28 Dr Genevieve Hayes
But what is the difference between a copyright or patent and a trademark?
00:08:32 Helen McFadzean
OK, so there are different types of IP protection.
00:08:37 Helen McFadzean
There are registered rights and there are things like copyright where you don't need to register, just subsists as soon as you know somebody creates a piece of work, I'll go to the register rights.
00:08:47 Helen McFadzean
So for patents, patents are for the protection of invention. So if it's something that's functional, if you invent something new in the area of science and technology, typically that would fall under patent protection. Trademarks are for brands. So you know things like Coca Cola.
00:09:08 Helen McFadzean
That will be a trademark.
00:09:10 Helen McFadzean
There's one more type of registered right, which is designs and that is to protect the visual appearance of something.
00:09:17 Helen McFadzean
So if you design a new watch face and you want to protect what that watch face looks like, then you could use a registered design to do that.
00:09:28 Dr Genevieve Hayes
Is this like how I think?
00:09:30 Dr Genevieve Hayes
Cadbury trademarked the shade of purple that they use on their wrappers.
00:09:35 Helen McFadzean
OK, so that's a trademark, not a design. So trademarks you can you can. There's different forms of trademarks you can have, you know, just the word mark. You can have a logo, you can have a word and the logo you can. There's a.
00:09:48 Helen McFadzean
Non traditional types of trademarks like colour, so you know a specific shade of purple for cadburys chocolate. The distinctive shade of you know, what would you call that colour for Tiffany greenish type.
00:10:01 Helen McFadzean
Robins egg blue. That's the one. You know the the specific shade of red at the the back of those shoes for what's the guys name? Chris Christie.
00:10:13 Helen McFadzean
That's the one. So you can have those types of trademarks as well, and you can have shape. So things like, you know, the Coca Cola bottle, the shape of the Coca Cola bottle could be a registered a, A trademark. You can also register the shape as a design.
00:10:34 Helen McFadzean
But a design has a limited time, so in Australia it's 10 years.
00:10:39 Helen McFadzean
But if you have it registered as trademark, A trademark can go on for indefinitely as long as you keep paying the renewal fees every 10 years.
00:10:48 Helen McFadzean
You can have the trademark registered for as long as along as long as you're you want, so that can be quite a a powerful a piece of IP protection. And so then for copyright copyright.
00:10:59 Helen McFadzean
Is more.
00:11:00 Helen McFadzean
For the expression of an idea, so it doesn't protect the idea itself, it's the specific expression of the idea.
00:11:07 Helen McFadzean
So things like written material, book, books, artistic works, cinematographic works, those types of things, software, code, the code itself would be.
00:11:20 Helen McFadzean
If you're protected.
00:11:21 Dr Genevieve Hayes
Oh, it's interesting cause I'd never thought of software code as being copyright. I'd always thought of things like books and movies.
00:11:28 Helen McFadzean
Includes includes the software code as well, so with something like the Windows operating system that would be subject to copyright, correct? Correct. You know, so proprietary code will be copyright protected and we'll use it under our licence.
00:11:44 Dr Genevieve Hayes
Right. That's very interesting.
00:11:46 Dr Genevieve Hayes
And what of our patents?
00:11:48 Helen McFadzean
Patents protect the functionality. It's quite a broad type of protection, so so for example, if we go back.
00:11:54 Helen McFadzean
To talking about software, the code itself will have copyright protection, but then you know that copyright can't stop someone from copying the functionality if they just look at how it functions and they.
00:12:06 Helen McFadzean
Go write their own code.
00:12:08 Helen McFadzean
But they're not copying the code itself, so if you want to protect the functionality of something really, the only way to do that will be through.
00:12:16 Dr Genevieve Hayes
Pattern protection. So if we take, I'm just thinking of my cell phone. So I've got an iPhone. So apple.
00:12:24 Dr Genevieve Hayes
And if you turn off a Apple iPhone, you've got the swipe right to turn off. Yeah, so that would be there'd be code sitting behind it that would cause that to work. But there's also the functionality that if I swipe right. I.
00:12:44 Dr Genevieve Hayes
Turn off my.
00:12:45 Dr Genevieve Hayes
Phone. So if Android were to put that in there is that functionality in the Android operating system?
00:12:55 Dr Genevieve Hayes
Even if that was using different code, if that was patent protected, they wouldn't be able to do.
00:13:01 Dr Genevieve Hayes
That is that right?
00:13:02 Helen McFadzean
They would be able to, if they have a licence from the pattern holder. OK yes. And with the electronics and particularly in the area of say for example, telecommunications, there are these things called standard essential patterns.
00:13:15 Helen McFadzean
So you have these patterns that form the standards so that you know things can talk to each other.
00:13:21 Helen McFadzean
And there are.
00:13:22 Helen McFadzean
Guidelines for licencing those so that it's fair for players in this area to use the technology under licence.
00:13:32 Dr Genevieve Hayes
Agreement so that everything is compatible with everything else.
00:13:35 Helen McFadzean
Correct. That's right.
00:13:36 Dr Genevieve Hayes
To avoid the situation where iPhones can only phone iPhones and Android phones can only phone Android phones and yeah.
00:13:44 Helen McFadzean
That would not be a good situation.
00:13:47 Dr Genevieve Hayes
I always thought you know how Tim Berners Lee, who invented the Internet, basically gave it to the world for free.
00:13:55 Dr Genevieve Hayes
I always thought, imagine how annoyed his family were, but he gave it to the world for freaks. They must be thinking, gosh, we could have.
00:14:02 Dr Genevieve Hayes
Been billionaires based on this?
00:14:05 Helen McFadzean
And I think you.
00:14:06 Helen McFadzean
Know that it's.
00:14:07 Helen McFadzean
Something to maybe think about is, you know, sometimes when you're working on new development, maybe IP is not real.
00:14:15 Helen McFadzean
The you know, at the front forefront of your mind. But you know, once you let the cat out of the bag, it's difficult to then retrospectively get protection if depending on where you've disclosed it.
00:14:26 Helen McFadzean
So it it's quite important to you know, have things like this so that you raise awareness in the R&D community, people are doing work to.
00:14:35 Helen McFadzean
To to solve problems and come up with new.
00:14:38 Helen McFadzean
Things that you know that they kind of think about IP protection concurrently with, you know, whatever their commercial goals might be so that they don't lose out on the opportunity to protect their IP if that is something that.
00:14:50 Helen McFadzean
They want to pursue.
00:14:51 Dr Genevieve Hayes
How could you potentially lose it?
00:14:53 Dr Genevieve Hayes
So would it be if I told the world I've came up with this great invention and then people started copying me and then?
00:15:00 Dr Genevieve Hayes
A month later I said Ohh. Actually I'd like to pay. I type it. I couldn't.
00:15:03 Helen McFadzean
Yes, well, so the.
00:15:06 Helen McFadzean
It it's based on, so there's three main.
00:15:09 Helen McFadzean
Requirements to getting valid patent protection and that part kind of is based on one of those requirements and one of the main requirements is that the invention needs to be new and so novel at the time of filing for your patent application, and so then it's incredibly important to keep everything confidential.
00:15:29 Helen McFadzean
Until you worked out, you've spoken to attorney, worked out what your patent strategy is, because if you go and publicly disclose your invention before you filed.
00:15:39 Helen McFadzean
Application then your patent application at the time that it's filed is not new anymore. In light of your own disclosure.
00:15:47 Dr Genevieve Hayes
What happens if you just tell a few family and friends?
00:15:50 Helen McFadzean
So yeah, so if it's, you know, a confidential discussion and everybody understands it's confidential, then that's probably fine. Still, I would advise to to keep it confidential until you've worked out what you.
00:16:04 Helen McFadzean
Wanna do there?
00:16:05 Helen McFadzean
Are provisions we've got in Australia, there's a 12.
00:16:08 Helen McFadzean
Month grace period provision.
00:16:09 Helen McFadzean
To help us salvage situations where you had inadvertently disclose and you still want to pursue patent protection, you just didn't know.
00:16:17 Helen McFadzean
But that's only for 12 months, so if you you've disclosed or you've written journal articles, you've published all of the details, and you've waited more than 12 months to go talk to a patent attorney.
00:16:29 Helen McFadzean
That's a little bit too late.
00:16:30 Helen McFadzean
Yeah, not much we can do at that point.
00:16:33 Dr Genevieve Hayes
So 12 months. OK, I'll keep that in mind.
00:16:35 Helen McFadzean
12 months not in Australia, so you know there's no such thing as a worldwide pattern so often organisations will want to protect their invention, say in our major trading partners at US Europe.
00:16:47 Helen McFadzean
So not every country around the world has this provision, so it's better to just not disclose. And if you absolutely, absolutely.
00:16:55 Helen McFadzean
Must and there's not enough time. Have at least you know an NDA. A non disclosure agreement in place to make sure that the any meetings that you have or any discussions you have are are confidential.
00:17:08 Helen McFadzean
But ideally you you.
00:17:09 Helen McFadzean
Should just file it before you have the.
00:17:12 Dr Genevieve Hayes
Would you have to file it in just so you wanted to cover the major jurisdictions so Australia, European Union, US, UK major Asian countries, things like that? So would you have to file all of those patents simultaneously?
00:17:29 Helen McFadzean
No. So how the?
00:17:31 Helen McFadzean
Process works, in a nutshell, is that you can file what's called a provisional application, and we have international treaties with most of our trading partners.
00:17:40 Helen McFadzean
So a provisional application will give you pending patent protection for most industrialised countries around the world for a period of 12 months.
00:17:48 Helen McFadzean
And that's just the the application that sits with the Australian Patent Office doesn't get published. It just doesn't to get examined, it just sits there and your rights, you have some rights that start from the priority date, which is the date that you file and then.
00:18:01 Helen McFadzean
12 months has gone by. Sorry. 12 months passes from the date that you filed. Then you can make some decisions as to how you want to continue with the pattern protection process. So you can either file like another bundle application which is called a an international PCT application.
00:18:18 Helen McFadzean
Holds kind of your rights pending in most industrialised countries for a further 18 roughly 18 months. Or you can then go into the specific countries that you want to directly say file that Australian complete application file a US application and then those applications eventually will be examined and become.
00:18:39 Helen McFadzean
Granted, the holding bundle application, the PCT application, just.
00:18:43 Helen McFadzean
Delays that process for a further 18 months, so at the end of the PCT phase, you still have to select the individual countries.
00:18:52 Helen McFadzean
It's just that you might not be depending on where the commercialization process is at. You might not be at a position to make that decision. As to you know, where you're going to be.
00:19:03 Helen McFadzean
Making the most money out of your invention, for example, you haven't, you know.
00:19:07 Helen McFadzean
Sorted out all of your investment deals and you know how how you're gonna proceed. So if you needed some more time, there's strategies that we can we can put in place to help you get some more time.
00:19:17 Helen McFadzean
Usually in terms of costs, when you start to select the individual countries to file them, there's a a large proportion of costs that will be payable at that stage, something to.
00:19:27 Helen McFadzean
Kind of get bear in mind is what the patent process is, that it's an ongoing cost you'll be paying cost at filing, paying costs at 12 months to complete eventually when everything gets examined, you'll be paying cost at the examination stage. Once it gets granted, there will be fees payable and then.
00:19:44 Helen McFadzean
Even after that renewals to keep it alive, so there's costs payable at stages throughout the life of the patent application and eventually when it becomes granted as a patent and it's staggered in a way to kind of so that you're not up for all of your cost at once, if you are also concurrently, ideally you're concurrently commercialising the technology.
00:20:05 Helen McFadzean
And somewhere along the process, you'll be making money out of it and.
00:20:08 Helen McFadzean
You can use a portion of your revenue to fund the IP.
00:20:13 Helen McFadzean
Costs, and if at any stage you decide that you know this is not something that's not worth pursuing anymore, you can just decide to abandon the application.
00:20:22 Dr Genevieve Hayes
Is there a point where you can no longer renew a patent so where it automatically goes into the public domain?
00:20:29 Helen McFadzean
Yes, so 20 years usually it's 20 years for pharmaceutical patents you there's possibility to extend that term slightly but for most so for for in our technology area software electronics it's it's 20 years so you can't extend it anymore after.
00:20:45 Dr Genevieve Hayes
That so I was just thinking with pharmaceuticals, so they.
00:20:49 Dr Genevieve Hayes
You can remain or you can keep on renewing those.
00:20:52 Dr Genevieve Hayes
But what about something like the secret in Greek? The secret recipe for Coca Cola is.
00:20:58 Helen McFadzean
Ohh, that's not a pattern. That's a trade secret and it's it's it's a secret for as long as you.
00:21:03 Dr Genevieve Hayes
Keep it a secret. So it's.
00:21:05 Helen McFadzean
OK. Yes, it's not a registered, right. Yeah, so you know, uh, Kentucky fried chicken. I think the kernel recipe. Yeah.
00:21:13 Helen McFadzean
The the herbs and spices or coca colas ingredients, so those I believe are trade secrets and it's not a registered right.
00:21:23 Helen McFadzean
You basically have to put controls it it it falls under the confidential information, so you need to put controls in place to make sure that it's recognised as a trade secret and that do keep it a secret.
00:21:34 Helen McFadzean
Because once, once the secret is out, then then it's not a trace secret.
00:21:37 Dr Genevieve Hayes
Anymore. So presumably anyone who works in whatever part of KFC repairs the kernels.
00:21:44 Dr Genevieve Hayes
Special blend of 11 herbs and spices. They would have to sign some sort of confidentiality agreement that they don't tell the world.
00:21:52 Helen McFadzean
Probably will be part of the employment agreement I presume if they do come into contact with that information. But if you're just working at KFC and you have this flower mix that's got the herbs in it, you unless you do some lab testing, you're not gonna know what's in there.
00:22:08
Have you ever.
00:22:08 Dr Genevieve Hayes
Googled what those 11 herbs and spices are.
00:22:11 Helen McFadzean
I have not have you.
00:22:13 Dr Genevieve Hayes
Yes, I have. I I don't know if this is real or not, but almost all of them are salt and pepper.
00:22:18 Helen McFadzean
OK, that's why tastes so.
00:22:20 Helen McFadzean
Good. Yeah.
00:22:22 Dr Genevieve Hayes
So it's like salt, black pepper, white pepper, garlic salt, MSG.
00:22:27 Helen McFadzean
Ohh all the good stuff, yes.
00:22:32 Dr Genevieve Hayes
I understand what you've said, but as someone who's never filed a patent before, if I came up with the next ChatGPT, I still probably wouldn't know whether it was worth protecting or not, because I just don't have experience in this field. Is there anythings? Any guidelines?
00:22:52 Dr Genevieve Hayes
You could give someone like me or one of our listeners to help us determine whether our brilliant invention is worth protecting using a patent or similar, or if it's just rubbish code that no one cares about. Well, OK, so I guess there's two a few different considerations.
00:23:09 Helen McFadzean
You know when you're developing something new, it's it's good to consider what your end goal is. So what is are you trying to commercialise this?
00:23:19 Helen McFadzean
Are you just trying to do something you know to share, to contribute to, you know, any any area to share knowledge and promote learning? If it if there's some kind of?
00:23:28 Helen McFadzean
Commercial objective. Then it will be useful to consider whether some form of IP protection, not necessarily patents, because not everything is suitable for patent protection.
00:23:38 Helen McFadzean
And I'll talk a little bit more about that in a minute. You know, where there's some form of IP protection will help you achieve that goal. So if you're.
00:23:46 Helen McFadzean
You know, releasing some kind of a product or you're trying to draw investment, you're trying to do a a deal with other business partners.
00:23:55 Helen McFadzean
It might be worth considering having some IP protection to give you more leverage in those negotiations. I mentioned that not everything is suitable for patent protection. I also mentioned that the the three main criteria.
00:24:06 Helen McFadzean
One of them was new, so you need to have, you know, come up with something that is new over everything that currently exist.
00:24:15 Helen McFadzean
And the second requirement is that whatever is new needs to be not obvious for somebody that has general understanding of this technical area.
00:24:27 Helen McFadzean
An example I could give, not necessarily in software is for example you, you you know, somebody invite invents a pen and you just invent.
00:24:35 Helen McFadzean
A different pen that writes in a different colour. You know that's not inventive enough, so there's those two criterias and there's a third criteria which is particularly important for computer implemented inventions, because there's this thing called patentable subject.
00:24:51 Helen McFadzean
Yeah. So not everything that exists under the sun can be protected by patents. Things like abstract ideas, business methods, those sorts of things are not patentable subject matter by themselves, and it's it's relevant for, you know, foundation technologies like software and.
00:25:11 Helen McFadzean
They are a machine learning because it can be these tools can be applied in any industry. So you've got computer implemented.
00:25:20 Helen McFadzean
Systems for you know, business administration in Fintech, E Commerce and online marketing, for example. And there it's just kind of raises the question of if business methods are not patentable by themselves, at what point will the computer implementation help push that over the line to meet that requirement for patentable subject matter?
00:25:40 Helen McFadzean
And this is a really complicated area of law, and it's been changing for the entire time that I've been in this profession. I remember, you know, in the beginning when I, when I first started.
00:25:51 Helen McFadzean
14 years ago it was quite easy to get computer implemented inventions over the line. You just had to kind of say that, you know, whatever software method that you're implementing is implemented on a computer.
00:26:04 Helen McFadzean
These days, the requirement is the bar is a lot higher. It's a lot harder to get computer implemented inventions.
00:26:11 Helen McFadzean
Over the line, in order to meet this pandal subject matter requirement, I guess to summarise the requirement, the key is that there needs to be something technical. So you need to be applying the machine learning or AI algorithm.
00:26:25 Helen McFadzean
Of them.
00:26:26 Helen McFadzean
To solve some technical problem you know there needs to be some technical purpose. It it could be for example applied to some technical problem that's outside the computer. For example, you know a control system that's controlling a robot arm manufacturing line.
00:26:45 Helen McFadzean
You're doing some image processing for obstacle detection, obstacle avoidance detection. You're using it in, you know, medical technologies, for example.
00:26:55 Dr Genevieve Hayes
Anything in a self driving car pretty much.
00:26:56 Helen McFadzean
Yeah, exactly, exactly the problem.
00:27:00 Helen McFadzean
Comes when you're using the machine learning AI algorithm to maybe just make just automate a business process, business administrative process, or you've got some.
00:27:13 Helen McFadzean
Yeah. So you've got something that might be a manual reporting process and you're just automating it on a computer and there might be some AI.
00:27:20 Helen McFadzean
Machine learning models in there to help make some predictions to make you determine I don't know something that's business related.
00:27:27 Helen McFadzean
So then you don't have that technical problem that you're solving outside the computer. The requirement then shifts to the consideration of the computer implementation. So are you making?
00:27:39 Helen McFadzean
In the implementation of that software code, are you doing something clever with, you know, memory allocation? Are you making the computer run more efficiently?
00:27:48 Helen McFadzean
The language that they use to test this is whether or not there's an improvement to the functions of the computer. That's irrespective of just the data being processed, and that can be quite a difficult requirement to meet.
00:28:01 Helen McFadzean
In real life, when you're solving these business problems or you're implementing the software solution for an E commerce solution, you're not really thinking about how to make the computer run more effectively or efficiently necessarily. Maybe because you're thinking about how to solve this business problem.
00:28:20 Helen McFadzean
But that's not to say that.
00:28:22 Helen McFadzean
Doesn't happen. It's just something to think about if you want to pursue pattern protection for something that is in those areas in those E commerce and and fintech areas that the focus will be the computer system rather than the the data it's processing to give you the benefit.
00:28:40 Dr Genevieve Hayes
OK, so if we use the example of some sort of.
00:28:43 Dr Genevieve Hayes
Stock Trading Company, a computer system where people can enter in their trades for the stock market that probably wouldn't be patentable because it's basically just an electronic version of what could have been done by a patient, a paper form.
00:29:01 Dr Genevieve Hayes
But if it was software that automated trades based on special criteria in order to maximise profits, that sounds like something that might be potentially patentable.
00:29:14 Helen McFadzean
If there is some improvement to the computer rather than, you know the business outcome, the financial outcome. So if you're, you know if there's the data sets so huge you've had to change how you process the data to make it viable.
00:29:29 Helen McFadzean
You know, usable in real life or in real time, so the the technical improvements are the computer system itself rather than rather than the the transformation of data.
00:29:41 Dr Genevieve Hayes
OK, so if it was because of this software, the computer runs faster and more efficiently in performing its trading that it can but actually just.
00:29:54 Dr Genevieve Hayes
Programming the system to know when to perform the trade. That wouldn't necessarily be patentable.
00:30:01 Helen McFadzean
Yeah, it might be difficult if you're just, if that's just a a business problem that you're you're solving that or or finance solution that you're providing the criteria that it needs to improve the function of the computer irrespective of the data.
00:30:16 Helen McFadzean
So if you think, OK, we're we're not, we're not using this to run financial data analysis where you can use it to run, I don't know something else, some other data.
00:30:24 Helen McFadzean
You still have that that improvement is still there regardless of what data you're feeding through the system. It's almost like saying so, you know, the application is relevant.
00:30:32 Helen McFadzean
And then you're actually just looking at the computer implementation. The software solution itself does something clever and and and solve some technical problems within the computer.
00:30:44 Helen McFadzean
You know, like with bandwidth, you know, efficiency, those sorts of things. So quite a difficult requirement to meet.
00:30:53 Helen McFadzean
In my experience, because typically in in real life you might not be thinking about doing that when you're implementing this office, but I I should mention that this is an area of law that is still changing.
00:31:06 Helen McFadzean
And actually in Australia, So what I mentioned to you is what, what the criteria, the Patent Office, the the examination practise of the the Australian Patent Office and these are the rules that they're giving us when we go through examination for these types of patent applications. But in terms of case law in this area.
00:31:26 Helen McFadzean
Where it's very unsettled.
00:31:28 Helen McFadzean
Was actually last year. We've had a a High Court case, which is the the highest authority in Australia. We had a split decision.
00:31:38 Helen McFadzean
So 6 judges heard the appeal and three judges said one thing and three judges said something else. So we're in, you know, the most probably as.
00:31:49 Helen McFadzean
Uncertain. That's it. It can be at the moment in Australia. Hopefully things will change. So I guess the only thing that we know we know for sure is that the scary of law is probably going to change.
00:32:00 Helen McFadzean
But we just don't know when. And at the moment this is what we know and this is what we know also based on what is happening in Europe and US.
00:32:09 Helen McFadzean
So usually when we draught a patent specification, we're not just drafting it for Australia, because often organisations will want protection in the US or Europe. So we kind of need to.
00:32:20 Helen McFadzean
Be mindful of all of these requirements that are in these other jurisdictions as well, so that when we're drafted we we make sure we have the best quality specification to give us the best fighting chance of meeting these criterias in other places.
00:32:35 Dr Genevieve Hayes
I'm still having a bit of trouble getting my head around how you could potentially or what a patentable AI or machine learning based invention might look like. Can you give me some examples so that I can visualise that?
00:32:50 Helen McFadzean
Yeah. So two things. Either you have a technical application for the machine learning. So in the area of like you mentioned before, in autonomous vehicles, you might be using it to solve a particular problem in a specific.
00:33:06 Helen McFadzean
A field of science and technology, those things are are usually quite easy to get over the line for patentable subject matter. It's only the E Commerce Fintech online marketing online bidding systems. For example, when you're using the.
00:33:26 Helen McFadzean
Software to solve a business problem. It's a it's a lot.
00:33:30 Helen McFadzean
Harder because we don't because.
00:33:32 Helen McFadzean
The business business methods are considered not patentable subject matter, so there's a for those solutions. The requirement for technical consideration in the computer implementation is a lot higher than it would be otherwise if you had some technical. If you could meet that technical requirement.
00:33:54 Helen McFadzean
Because you're solving something, you know, you you're you're working in the field. That is technical by nature.
00:34:01 Helen McFadzean
Yeah, so it's.
00:34:02 Dr Genevieve Hayes
Basically, if it's an engineering application of machine learning or AI, then you're probably good. Yep, but if it's a.
00:34:08 Helen McFadzean
The office worker business type thing. Then you may run into problems. Yes. Yes, that's right. If it's a, you know, a new marketing tool or an app to to, to to change consumer behaviour. So. But it really depends on some. It's very blurry because it depends on what you're doing.
00:34:28 Helen McFadzean
If you're doing, if you're, if you're. If you're still solving some technical problems in the computer or in in your in your business application, you might be doing something clever with.
00:34:38 Helen McFadzean
I don't know image processing, maybe for example will be a good example to determine consumer behaviour. Maybe there's some science behind?
00:34:48 Helen McFadzean
That that might be helpful. What?
00:34:50 Dr Genevieve Hayes
About in something like medical areas. So I'm thinking I remember watching this video on the Internet about this person who had created a machine learning model that was able to take a photograph of some sort of skin lesion and.
00:35:08 Dr Genevieve Hayes
Classified as either malignant or benign.
00:35:11 Dr Genevieve Hayes
And obviously something like this would have a lot of benefit as a diagnostic tool for dermatologists. Would something like that potentially be patentable? Yes, I would say so. You still have to meet the requirement for novelty and inventiveness.
00:35:29 Dr Genevieve Hayes
And the fact that I've seen a YouTube video and it tells you it's not novel or inventive.
00:35:34 Helen McFadzean
They might have had protection before they release.
00:35:36 Helen McFadzean
So yes, so you you you, that's why it's important to kind of think about that.
00:35:40 Helen McFadzean
Before you go and start promoting it and telling the world about your invention to have kind of your IP.
00:35:46 Helen McFadzean
Strategy already sorted out. Stay away from social media. Basically. Yeah. Don't tell the world about your invention just yet, yeah.
00:35:55 Dr Genevieve Hayes
With a machine learning model, you have all the different parts of a machine learning model, so you've got your training data, the algorithm, that's.
00:36:06 Dr Genevieve Hayes
You used to train the model and your final model fitted model itself. Which part of the model are you actually patenting? Is it the final fitted model or all those bits that went?
00:36:18 Dr Genevieve Hayes
Into it.
00:36:19 Helen McFadzean
That's a really good question. OK, we need to talk about patentable subject matter again.
00:36:25 Helen McFadzean
So mathematical algorithms. So the simplest form example I could give you.
00:36:31 Helen McFadzean
You know a mathematical equation F equals MA equals MC squared. The equation themselves are not considered patentable subject matter.
00:36:40 Helen McFadzean
So the models are machine learning model, so the European Patent Office considers a machine learning model itself just the maths itself. That's not patentable subject matter because that's just abstract maths.
00:36:54 Helen McFadzean
It's just a complex version of say F = m Eight. That's the how they categorise it. But then if you are training a model for a technical purpose then that's OK.
00:37:03 Helen McFadzean
We're going back to the technical application again. Say if you developed a new machine learning model that you could just use for anything, you know, say you were the first person to develop the neural network, a basic neural network.
00:37:15 Helen McFadzean
You couldn't just patent the the the the neural network itself.
00:37:20 Dr Genevieve Hayes
So the algorithm.
00:37:21 Helen McFadzean
The model itself, but you could if you have used it to solve some technical problem, applied it to for a technical purpose, and that potentially could be patentable in terms of what we're actually claiming.
00:37:33 Helen McFadzean
So maybe I need to go back and explain what a patent specification looks like, so it's kind of a a pattern.
00:37:40 Helen McFadzean
So in order.
00:37:41 Helen McFadzean
To file a patent application you have to prepare this thing called a patent specification, and it's quite a lengthy document.
00:37:46 Helen McFadzean
It's part technical, part legal. You have the first part of the application which just talks about the problem, the.
00:37:54 Helen McFadzean
And it is described the invention in sufficient detail for someone else to who understands this field of technology to be able to put it into practise without having to do any additional inventing by themselves.
00:38:07 Helen McFadzean
So that's forms the description part and then at the end of your specification you've got a list of paragraphs.
00:38:13 Helen McFadzean
Numbered paragraphs called the claims, and that's the legal part. The most legal part of the document that defines the scope of protection and that needs to be supported by your description. So how we could craft the claims for a computer implemented invention involving machine learning model is that.
00:38:32 Helen McFadzean
You could claim so software typically are all method claims because software is just executing you know instructions. It's just it's executing method essentially.
00:38:43 Helen McFadzean
So we can we can claim the the training method to to for our technical purpose. So it needs to be training it to do something and we can also claim.
00:38:54 Helen McFadzean
A system you know that might include some hardware parts and the.
00:38:58 Helen McFadzean
Software part. So if I'm thinking of a system for autonomous drive or for determining how to move in an autonomous vehicle, it might include some hardware sensors.
00:39:08 Helen McFadzean
It might include the data from the sensors. It might include a machine learning model that reads the sensors and freeze the data through something, or it's already a learned model. A trained model that it just spits out the decision.
00:39:19 Helen McFadzean
And then tells the the vehicle what to do.
00:39:23 Helen McFadzean
So that we can we can protect that as an overall system, we can also claim some hardware components that has the trained model saved on it.
00:39:34 Helen McFadzean
So you know, we would typically define what's it called, you know, a a computer readable medium. Basically you know the memory part of the computer.
00:39:42 Helen McFadzean
And you can you can. You can claim that which contains the model trained in accordance with, you know, a method that you've you've set out which is the training method in terms of the protection pattern. Protection is quite broad ranging. So if you have claims to the method.
00:39:59 Helen McFadzean
You would also have some protection to the product which results directly from the method. So if you claim the training method, the direct product of that training method will be the trained model, I think you'll likely also have some protection for that train model as well, but you can also claim directly the.
00:40:19 Helen McFadzean
Hardware which has the model saved on it, something I should also mention is that because this area is so new.
00:40:26 Helen McFadzean
New we've seen exponential growth in the number of applications that involve machine learning and AI solutions in, say, like the last, I don't know, 5-7 years.
00:40:37 Helen McFadzean
So these applications are still going through examination processes and we don't have a lot of case law in this area, so we.
00:40:46 Helen McFadzean
Don't know. We don't have a lot of answers for certain, but we suspect you know moving forward.
00:40:53 Helen McFadzean
Some of these will be litigated and we'll know more, but in the meantime, you know, we just. Yeah, so we we're.
00:41:00 Helen McFadzean
This is what what we're we're doing is trying to craft the specification a way that gives you the best protection from all of the different aspects.
00:41:08 Dr Genevieve Hayes
When you're describing the pattern application before you were mentioning the description.
00:41:13 Dr Genevieve Hayes
Now I can imagine that would be relatively easy to do if you were describing a physical object, because you could say you know it's this size. It looks like this. It's got these bits and pieces inside it.
00:41:25 Helen McFadzean
Does this how do you describe something which only exists in the form of computer software? Really, that's a really good question, and it it's a challenge for this particular area, you know, because it's not something that you can see. You can't just look at a CAD drawing and figure out how it works.
00:41:47 Helen McFadzean
The preparation of good quality pattern specification is a collaborative effort between the patent attorney and the inventors, so communication is so important to make sure that.
00:41:58 Helen McFadzean
All of the relevant detail ends up in the pattern specifications, so you want to be working with the partner turning that understands this field of technology.
00:42:06 Helen McFadzean
So they're asking the right questions to get that information from the inventors. It it can be a little bit of an iterative process, particularly for people that don't have experience in this area.
00:42:18 Helen McFadzean
It's difficult to kind of know.
00:42:20 Helen McFadzean
Or what type of information you need to provide to a patent attorney because you know your.
00:42:26 Helen McFadzean
Project could be.
00:42:27 Helen McFadzean
Huge and you might not have documented at all and you know how? Where do you even start?
00:42:31 Helen McFadzean
Right. So a lot of that you need to have guidance from the patent attorney to kind of get the information from you, it might go back and forth a few times.
00:42:41 Helen McFadzean
Until you have all of the relevant information that you can.
00:42:44 Helen McFadzean
Used to then and support a good quality specification, but it can take some time depending on how much of that information is readily available and how much you kind of have to go back and prepare.
00:42:54 Helen McFadzean
Probably looking at, say, at least three to four weeks, maybe more to kind of put something like this together. You were talking about TV shows.
00:43:04 Helen McFadzean
Before and I remember watching an episode of Suits. Have you seen suits?
00:43:09 Dr Genevieve Hayes
No, Meghan Markle put me off it. Ohh.
00:43:12 Helen McFadzean
I watched it before she became.
00:43:16 Helen McFadzean
Yeah. Yeah. Became royalty. Yes, that's right. And I remember there was this episode in suits. It was one of the earliest ones and they it was about patents and and I think.
00:43:27 Dr Genevieve Hayes
All right.
00:43:31 Helen McFadzean
I vaguely remember what happened. It was there was an inventor and he went and had like, a investment. A meeting with some.
00:43:37 Helen McFadzean
Masters in the morning and they wanted to file a patent application on the same day and then so they were trying to work out how to file this patent application.
00:43:48 Helen McFadzean
They filed a patent application a day late and there was an issue with an earlier filed application, so they had to file a challenge to that earlier application. They.
00:43:57 Helen McFadzean
Eventually sorted out all of the issues and secured investment from the investors in the order of hundreds of 1,000,000.
00:44:06 Helen McFadzean
The dollars and all of that happened in like 2 days. It was just the most ridiculous thing. Ohh and I think the patent application got examined like as soon as it was filed.
00:44:17 Helen McFadzean
And you know, you had time to go play tennis during the day. It was so, you know, for for a patent attorney that that knows how long these things take you just.
00:44:27 Helen McFadzean
I don't even.
00:44:29 Helen McFadzean
It's it's just ridiculous on so many levels. I can't even begin to tell you.
00:44:33 Helen McFadzean
Yeah. So it usually you wouldn't, you would, if I look back at the last 14 years of my career, I think I only recall.
00:44:40 Helen McFadzean
So there's only been one time where I've had to file a patent, prepare patent specification from scratch and file it on the same day because the client wanted to disclose the next day, we had the information the day before and that was a.
00:44:52 Helen McFadzean
A frantic day, I can tell you I did not have time to.
00:44:56 Helen McFadzean
Go play tennis during the day.
00:44:58 Dr Genevieve Hayes
Well, how long does it normally take?
00:45:00 Helen McFadzean
So normally it takes about three to four weeks if if the information is. If you have all of the information.
00:45:07 Helen McFadzean
At the start of the three to four weeks, the process of preparing the specification, it's it can be quite a lengthy document in our area software area, probably at least 30 to 40 pages for the simplest one I would say typically and the patent attorney would prepare that from scratch.
00:45:22 Helen McFadzean
It's not. We don't have templates or precedents for for it. We we, we we prepare it from scratch.
00:45:27 Helen McFadzean
So we need to allow time for an initial meeting to discuss what the invention is about to make sure the pattern.
00:45:35 Helen McFadzean
Dance, the patent attorney would then go away and prepare some claims. You might discuss that and make sure everybody's aligned, and then they go away to prepare the rest of specification that could.
00:45:46 Helen McFadzean
That could take about it a week or two, and then there's review processes. Depending on the size of the organisation and the size of the team and the number of people involved.
00:45:55 Helen McFadzean
Need to give people enough time to review and comment that needs to come back and you might, you know, have a a couple of.
00:46:01 Helen McFadzean
Rounds of revision, so all up.
00:46:03 Helen McFadzean
Probably come, you know, three to four weeks typically before the specification can be finalised and everyone's happy with it and you proceed with filing. But if you don't have all of the information ready to go at the outset and you're having an exploratory discussion with the attorney to figure out whether you know you've got something to patent and then you go away.
00:46:23 Helen McFadzean
And compile the information. You know that you need to give yourself enough time to put the put the put the information together and it it it's not.
00:46:31 Helen McFadzean
You know, we're not just filling in a form and ticking some boxes. It's quite involved process that that we need.
00:46:36 Helen McFadzean
To we need to explain in enough detail for someone else to put the invention into practise and give examples and describe the best method of putting the invention into practise.
00:46:47 Helen McFadzean
So there's quite a bit of information that we need to put into our specification and it's so important, so, so important to make sure that it's.
00:46:56 Helen McFadzean
Of good quality at the time of filing, because there's limited opportunities to.
00:47:03 Helen McFadzean
You fix any problems later on once you filed, and so much of it turns on the specification, so you know if you have a good specification, it's easier to to prosecute through the different pattern offices. It's easier to overcome those objections of novelty and inventive step and patentable.
00:47:23 Helen McFadzean
Subject matter. If you have, you know more information to work with.
00:47:27 Helen McFadzean
And you've just described the invention properly and also you know it's more likely to hold up to third party.
00:47:32 Helen McFadzean
Challenges so that you can enforce the IP right when you need to.
00:47:37 Dr Genevieve Hayes
What happens if one of the inventors isn't human? So if you created an invention with the assistance of some sort of?
00:47:47 Dr Genevieve Hayes
Generative AI tool like chat, UPT.
00:47:50 Helen McFadzean
That's an interesting question. So you might have heard of the dabas cases. I'm not sure if you have it's.
00:47:59 Helen McFadzean
It's it's very rare that anything in patent law makes it into mainstream media is one of the examples. So someone came up with this, this, this, this AI machine called dabos and allegedly solved some problems by itself and generated.
00:48:19 Helen McFadzean
These outputs that it considered inventions and and and that was filed into patent applications and filed around the world in in many different countries and that was litigated through the courts in Australia to determine purely to determine the question of whether.
00:48:35 Helen McFadzean
Not the word inventor in our legislation can be interpreted to include an AI machine, and the answer. The short answer is no.
00:48:47 Helen McFadzean
We need to have a a human inventor named on a patent application for the application to be valid. That's quite an academic.
00:48:55 Helen McFadzean
Question though because it's you.
00:48:57 Helen McFadzean
So the IT doesn't go into exploring whether or not you know in our current level of development whether or not in real life, if you have a Gai tool that facilitates the R&D process, whether or not that tool is actually an inventor because you know.
00:49:17 Helen McFadzean
The AI is just processing data and it's, you know, being trained. There's always a a person behind it that either built the AI model is training it setting.
00:49:27 Helen McFadzean
Matters and determining whether or not the output is viable, right? So it might help reduce the time of R&D because it's able to churn through more data and provide you with, you know, maybe some options to help you think of things that you you might not have.
00:49:47 Helen McFadzean
Otherwise, in your R&D process, but the person still needs to maybe determine whether or not that outcome is workable and potentially select one of many different outcomes because.
00:50:01 Helen McFadzean
You know or or modify in some way so that you maximise the the benefits or it it it it works better than than the other options that it provided you.
00:50:12 Helen McFadzean
So really, is the AI and inventor in that sense? I think what my personal opinion is that it's it's more of a tool that the inventor.
00:50:21 Helen McFadzean
Has used to help him or her or the team to come up with the invention rather than it qualifying as an inventor really cause it has no consciousness.
00:50:32 Helen McFadzean
It doesn't know what it's doing, it's just processing data and giving you some output. You know, it's like you asked the question, who discovered Jupiter's moon?
00:50:41 Helen McFadzean
It's it's. It's Galileo, not not his telescope. It just so happens that he also invented the telescope.
00:50:49 Dr Genevieve Hayes
Yes. OK. That makes sense.
00:50:52 Dr Genevieve Hayes
So is there anything on your radar in the AI data and analytics space that you think is going to become important in the next three to five years?
00:51:01
OK.
00:51:01 Helen McFadzean
I think it's such a revolutionary and powerful foundation technology that can be applied to so many different industries. I think generally we're we're going to see wider adoption. It's just gonna be in. I mean, it's already in so many aspects of our lives.
00:51:18 Helen McFadzean
But maybe more so in the next three to five years. I think for me, I would like to see it help us, you know, solve some of the the big problems that we are facing like for example, climate change, you can call me cynical.
00:51:32 Helen McFadzean
Maybe I feel like a problem like that. You know, you you you need exponential shift in human behaviour.
00:51:40 Helen McFadzean
And or exponential shift in technology development. But the human behaviour piece is difficult if we're just relying on governments and policy change to slowly drive that it's such an urgent problem. We need to have a quicker solution.
00:51:56 Helen McFadzean
And they I potentially could help us, you know, come up with the solution and you know, so I I think I I don't know, I don't have for crystal ball I I I hope that in the next two to five years we can have some more AI driven or technology driven innovations to you know really help us tackle some of these big problems that we're facing.
00:52:17 Helen McFadzean
For humanity, for our planet.
00:52:19 Dr Genevieve Hayes
I reckon it's got a lot of potential in the medical space, so developing cures for horrible diseases.
00:52:26 Helen McFadzean
Yes, yes, yeah.
00:52:27 Dr Genevieve Hayes
Yeah, that's that's definitely something I that's definitely a major world problem. I think it could solve in the next well.
00:52:35 Dr Genevieve Hayes
Everything takes longer than you expect, so I'd like to say three to five years, but at some point anyway.
00:52:42 Helen McFadzean
Yes, yes. Good to have the tools to make it possible. So it's it's exciting that you know I I I know AI comes from a a long history and it's only really in the last say 1020 years that it's really taken off with you know having more powerful computers and having the data sets there so.
00:53:00 Helen McFadzean
Even though it's it's taken off, I feel like we're still kind of just at the we're at the we're, we're just scratching the surface. You know there there's so much more potential. So who knows it you know.
00:53:13 Dr Genevieve Hayes
What final advice would you give to data scientists looking to create business value from data?
00:53:19 Helen McFadzean
I think my main advice would be to think about the commercial strategy early and if you're interested in IP.
00:53:28 Helen McFadzean
And talk to a patent attorney as soon as possible and keep your work confidential in the meantime. I can't tell you how many times I've had meetings where the inventors are so excited about what they've developed, but then halfway through the meeting, they tell you they've disclosed it more than 12 months ago and it's it's heartbreaking because there's nothing. You know, there's nothing anyone can do at that.
00:53:49 Helen McFadzean
At that point, if it's already been disclosed, it's already out there this within 12 months, maybe there's something we could do.
00:53:55 Helen McFadzean
It's not ideal, so I think the most important thing is to to think about it early and to get the advice early so that you don't lose that opportunity.
00:54:04 Helen McFadzean
Andy, just because you didn't know.
00:54:07 Dr Genevieve Hayes
Yes. So stay off social media.
00:54:11 Dr Genevieve Hayes
And for listeners who want to learn more about you or get in contact, what can they do?
00:54:17 Helen McFadzean
Happy for anyone to drop me an email or connect with me on LinkedIn. Send me a message. I'm always happy to have.
00:54:23 Dr Genevieve Hayes
A chat. OK. And I'll put your LinkedIn page up on the show notes.
00:54:28 Helen McFadzean
Great. Thank you so much.
00:54:30 Dr Genevieve Hayes
Thank you for joining me today. Thank you.
00:54:33 Dr Genevieve Hayes
And for those in the audience, thank you for listening. I'm doctor Genevieve Hayes, and this has been value driven data science brought to you by Genevieve Hayes Consulting.