A Patent Pioneer On Barriers, Breakthroughs, And The Next Wave Of AI - Virginia Driver

Virginia Driver recounts entering a profession that once enforced skirts and skepticism, then earning credibility across computing, telecoms, and the smartphone wars, where old devices became courtroom gold. That history matters: it shaped her insistence that AI must be auditable and accurate, not just fast. We unpack the real sweet spot for AI in patents—deep analysis across thousands of documents, transparent claim mapping, and outputs a human expert can verify. No hype, no black boxes, and a clear line from machine reasoning to attorney judgment.

Enjoyed the show? Subscribe, share it with a colleague, and leave a quick review so others can find us. What part of patent work do you most want AI to transform next?

Transcript

Cold Open And Shed Story

SPEAKER_05 0:00

Hey Gullum, we're um early into the next run of the next season of the podcast. Are you uh are you back into the swing of things, do you think?

SPEAKER_04 0:07

I um I've got a massive list of things to be rude to you about. It's going well.

SPEAKER_05 0:11

Excellent. Are you gonna do them all now or are you gonna save them up for a successive podcast?

SPEAKER_04 0:14

Oh no, I'm gonna it's come run and run, but you have got tons. It's great.

SPEAKER_05 0:18

Excellent, excellent. I'm I'm terrified because I I know that we've been together in various places that have been quite social, which might mean that I've been too liberal with stories.

SPEAKER_04 0:26

You are no totally way too liberal with your stories. You need to think. It's great. I just I if you've noticed, I don't know how you respond to them. I'll just stop typing into my phone when you go.

SPEAKER_05 0:36

Oh, yeah, yeah, yeah, yeah. Yeah, you stop talking to me and start taking notes. Yeah, I've noticed.

SPEAKER_04 0:39

I think today I want to talk about your shed.

SPEAKER_05 0:42

My shed. Okay, yeah, let's go for my shed.

SPEAKER_04 0:45

Uh and I want to give us some guidance about what to do. Which bits of the shed should you visit? Which bits should you perhaps avoid?

SPEAKER_05 0:51

So you're talking about me falling off the roof of my shed, aren't you? That gets right to the kind of the nuts of things. Yeah. Yeah. And I need to be very careful about how much I say, because this is all to do with a house move and just trying to get the house in the best possible shape for the new people moving in. Yes. And I thought it might be quite a nice idea at late o'clock at night when it was getting dark to jump up on my shed roof, uh, which is a kind of a uh not my shed roof, it's a house in the past now. It's a house that I've moved away from. And by shed, it's quite a large workshop with next to it what was quite a large chicken coop. But we got rid of the chickens years ago because we got too many fox strikes.

SPEAKER_04 1:25

I'm gonna add this to the pancer. Hang on, this is a separate pancer, these chickens, but period.

SPEAKER_05 1:30

So so I'm saying this because the shed has got actually it was quite a large workshop, has got a corrugated, uh it's not plastic, but it's one of those kind of Corollax type roof things. Whereas the chicken pen had sort of like a transparent plasticky roof on it. I don't know what it's called. So, yes, so I was just doing a little bit of late night tidying in the dark on my shed roof, and then and then needed to get down from the shed roof, and in doing so, rested myself on the plasticky bit and fell through. So fat fat so fat found found myself in the dark, hurtling towards the ground, tried to tried to grab the ladder as I went past it and only succeeded in pulling it down on top of me.

SPEAKER_04 2:12

There's lots and lots of questions about why you needed to be on the roof of your shed to do tidying, why you decided to do it in the middle of the night, various other things as well. But I as you say, you don't want to give it all away.

SPEAKER_05 2:24

No, that there may have been things in the vicinity of the shed that shouldn't have been there that needed to be removed, and it may have been best to do it under cover of darkness. So I'll just leave it at that.

SPEAKER_04 2:34

As long as it wasn't the chickens, we're all good.

SPEAKER_05 2:36

No, it wasn't the chickens, yeah. And I've not disposed of any bodies or anything like that. Uh I just didn't want the neighbours to see me doing something that looked a bit silly.

SPEAKER_04 2:50

I mean we don't want the podcast to turn into what Lee did next, but maybe we should. I should just call it that for a while.

SPEAKER_05 2:56

Have you never done this then? Have you never like fallen off of a roof? It's not the first time for me. That wasn't my first radio. How many roofs have you fallen off? Oh, several. And scaffolds. Really?

SPEAKER_04 3:07

It's just about harnesses and things and stuff like that.

SPEAKER_05 3:10

Oh, this was back in the days before health and safety. And again, and again, involved a house, my house, where I was on the roof, and had left the trapdoor, had left the trapdoor open in the scaffold. So I was climbing from the roof down onto the scaffold, not realising realizing I'd left the trapdoor open and fell straight down through the middle of it, two stories.

SPEAKER_04 3:27

I'll stop laughing and say, Were you okay?

SPEAKER_05 3:29

Yeah, I'm fine. Good, good, good. Good lord. Anyway, we should we should crack on with the podcast because I'm always concerned that we've got guests sat waiting in the in the ether ready to join us. And they might not want to hear you and I just talk in this sort of bant's way. Oh, actually, we did try and work at what the German for Bance was, didn't we, while we were away in Munich recently, but maybe we'll save that for another podcast.

Guest Introduction: Virginia Driver

SPEAKER_04 3:50

Oh, I'll write it, I'll write it down. I'll write it.

SPEAKER_00 3:53

Lou Davis and Willem Roberts are the two IPs in a pod, and you are listening to a podcast on intellectual property brought to you by the Chartered Institute of Patent Attorneys.

SPEAKER_05 4:11

No um exciting podcast today. We've got Virginia Driver with us. Virginia, hello, hi, welcome to the podcast. How are you?

SPEAKER_01 4:17

Hi, very well, thanks. Hi Lee, hi Gwyn.

SPEAKER_05 4:20

Have you wondered, are you wondering to yourself, what have I let myself in for?

SPEAKER_01 4:23

Uh not not not yet. Not yet, but it wouldn't it wouldn't be the first occasion when I've been uh doing something with Gwillem where that feeling has come across me later on.

Starting Out As A Woman In IP

SPEAKER_05 4:35

Oh, oh, Bance, Bance! Oh god. Well I I may have a closing question, Gillem. Uh so Virginia, would you do us the um honour of introducing yourself to our listeners, if you wouldn't mind?

SPEAKER_01 4:50

Yeah, so I'm uh Virginia Driver, European patent attorney of 30 plus years experience. I graduated with a degree in engineering science from from Oxford all those decades ago, went straight into becoming a European patent attorney, and I've spent my career in computer hardware and software. I worked for the firm of Page White and Farr until last June, built up their AI specialist patent attorney team for them, and and and now I have uh co-founded a tech startup bringing Gen AI to the patent profession.

SPEAKER_03 5:23

Cool. I'm gonna jump straight in because I know go for it. I know I know you're desperate to go for it.

SPEAKER_04 5:29

Are you I are you allowed to do you okay to say when you joined? Because I'm not quite sure whether you predate me in the profession, how long you've been doing this for.

SPEAKER_01 5:38

Yes, I started in the I started in the profession in I graduated in 1982, that in June, and started in the profession in September 82. I think I pre-date you, Willem, by potentially Sam Martin.

SPEAKER_04 5:55

But that's that's really interesting. So I'm I'm gonna guess that it was a slightly less common thing to see women coming into the profession back in those days.

SPEAKER_01 6:03

Well, yes, and and and probably like nearly every patent attorney of my generation, I came into the profession knowing virtually nothing about it. So I came into the profession on the back of a small leaflet that was handed to me by the careers officer at Oxford University who was struggling to place a female with engineering in into anything at all, and particularly a female whose practical skills had proved to be less than optimum for a career in in engineering. So for these reasons, I ended up uh in the world of in the world of patent attorneys. Uh, I was the first female hire that the firm I started with uh had made, and they made it extremely clear to me after a week that they'd only hired me because I was the only candidate. Ideally, they would have hired a man, but there wasn't one. That was my uh my intro. That was back it back in the day. That's how the world, that's how the world was.

SPEAKER_04 6:57

Oh well, I'm I'm sorry to hear that. Hopefully you made them realize that they made the best decision ever.

SPEAKER_01 7:03

Uh well I I'd like to I'd I'd like to I'd like to think so. I'd like to think so.

SPEAKER_04 7:11

So I'd like to I mean it's pretty quite difficult to imagine what the environment was like. I mean given how kind of the efforts for diversity and inclusivity and everything these days, presumably, did you did you find ceilings everywhere you went? How difficult was it?

SPEAKER_01 7:24

Uh the the the the when I started in the profession, ceilings, later other eyes were a long way off, of course. It was more a set of practical difficulties. We were required to wear skirts and smart shoes in the office. We couldn't wear trousers, which given that I cycled to work every day, you know, presented some sartorial challenge challenges to have skirts and shoes sufficient for sufficient for office wear. Yeah, the glass ceilings came came a bit later, uh, later on in my career, I would say. Uh I suppose a little bit sadly. I mean, the reason I left my first firm was because there was no prospect whatsoever of a female being promoted into partnership in in the late 80s uh at that time in that firm.

SPEAKER_04 8:12

Which is what we'd now describe as illegal.

SPEAKER_01 8:14

Uh which we would count now as as as illegal, yes, yeah.

SPEAKER_04 8:18

And um, well that with clients, presumably they also went into glassy shock when a female engineer turned up to do them anything.

SPEAKER_01 8:26

Um yes, yes. The actual engineers themselves were mostly pragmatic enough to speak to anyone who within five minutes demonstrated they could understand their language. But the see the senior executives and the business managers that that that took a bit longer. And in fact, for many years I just side hustled along with the senior male partner of the firm just in order to make sure those those doors were open. It wasn't unheard of for us to have a meeting and for people to come in and assume that I was the secretary waiting for the real guy to start the meeting. Uh and actually, I mean, people forget how quickly times have changed, but I would say that that persisted well into the night well into the 90s. And even the early 2000s, uh, the growth the actual assumption that you weren't probably the real deal if you were um particularly young, a young female professional. Yeah.

SPEAKER_04 9:31

Well, but thank you for forming the vanguard, because I think now it's it's such a the gender balance, well at least at the entry level is so so good now, so strong.

Early EPO Era And Career Path

SPEAKER_05 9:41

I think do you want to say a little bit about that, Gwillem? Because I know that I did some I did some stat work, stats work a few years ago when we talked on the podcast about the the kind of the pipeline around this. So we are now at about so when I joined CEPA back in 2012, I think the gender balance was sort of 25, 75. We're now at 38% of our members are female. There are more year on year more women enter the profession than men now. Uh so that there are always more uh new women trainees than there are men trainees. But I still did work out that it's going to be sort of like the mid-2040s at our current rate before we get anywhere near parity. Okay.

SPEAKER_01 10:19

And as a matter of interest, if I if I if I'm allowed, if I'm allowed to ask a question, given I've got you here, is how how does that break down in terms of technology sectors? Do you know how many of that is telecoms, compute, electronics versus biopharma, for example?

SPEAKER_05 10:35

I don't, Virginia, off the top of my head. I I could find out quite easily. So that's uh yeah, I'll I'll do that. I'll I'll feed I'll feed that back to you, but we'll also use that in a future podcast. Yeah, that would be interesting.

SPEAKER_01 10:45

Well, I think it is interesting, but just a little shout out for one of the things that I thank you for the appreciation, Gwim. It is appreciated because some things have been hard. I I said I spent the last decade or so building up an AI specialist team at Page White and Farrow where I hired computer scientists or mathematicians or quantum specialists and turned them into patent attorneys. And that team was 50% female when I left. And that was still pushing against a lot of headwind to achieve that. I'm actually I'm actually really proud of that.

SPEAKER_04 11:15

No, it's lovely. Well done. Actually, just one other interesting point. When you entered the profession, it was a cause of a bit of a huge flux because the EPO had started four years earlier. And what I hear is everyone thought, well, that's it for the UK profession, and everything's going to Germany. Well, so you've probably got an unusual hire. I think a lot of them weren't hiring at all around that.

SPEAKER_01 11:40

Well, uh, yes. So in addition to being, in addition to being the only the only female, the age profile was interesting because the next youngest patent attorney in the firm was 10 years older than I was. So I was so I was all alone as a as a single female trainee with a with a 10-year age gap to the next person, who was extremely nice. I can't I can't fault him for trying.

SPEAKER_04 12:02

And then were you handling European work back then? Or was it something that your old firm was still nervous about?

SPEAKER_01 12:09

No, no, they were that they were um they they'd done a pretty good job actually. Uh so I I I was doing a lot of UK uh you remember this, uh Gwillam, even when you started. There used to be a lot of translation work came in for uh ratification work. That's where I started uh my career was in in reviewing translations, but they did have a significant amount of European prosecution actually, even then. So they'd done quite well. That's interesting.

SPEAKER_04 12:33

Yeah, I was trying to find out what was going on in that in the 80s because I joined, I think, as the end kind of when when things start to look quite rosy and eco start to make a lot of sense as far as I can tell. That's interesting. And then I know that um you find yourself in telecoms big time, didn't you?

From Hardware To Telecoms And Litigation

SPEAKER_01 12:49

Uh I did, yes. In fact, I found myself first in computer hardware, which was a company called Inmos, which was one of the early British startups that looked like they might might be actually successful. And they developed a product called the Transputer, which was actually the first, the world's first attempt at parallel computing. So I was unbelievably lucky to to hit computers very, very early on. And that took me into other other compute hardware and software, and that took me into telecoms for a long time. And then that took me into other comms, peer-to-peer technology, you know, early days of of of packet-based telecom, IP-based telecom communication. I'm being very careful not to name names here because I don't know how. Yeah, exactly. And then and then morphed uh morphed from that via games actually. I did quite a lot of work in games for for a decade or so, and then into AI. So I got into AI early relative to many patent firms, actually, uh, probably about 15 years ago, started looking at it as the next big wave of technology. I couldn't possibly have known then how right that was how it turned out to be.

SPEAKER_04 14:06

We've talked about AI a lot, and I think so. We'll come to that in a in a sec, actually. And I I mean, I I definitely know one of your big clients, but if you don't, if you're not going to name them, I'm not going to either. Yeah. Yeah, but I mean the tell the I mean the the telecom story just quickly. I mean, I think you were very you must have been very central to a huge explosion in patenting in that area.

SPEAKER_01 14:24

Uh yes, no, very, very, very much so. I mean, we we we we acquired, yeah, very, a very big client in the scene and were preparing a huge amount of their of their portfolio. And then, of course, we went through the the sort of the smartphone wars era where everybody basically everybody sued everybody else for a period of about three to seven years, which was which for a panel attorney was was extremely exciting stuff actually, because although we weren't doing actual litigation, we were called in to support and and provide evidence about really, really old bits of kit, you know, the first sort of mobile product that purported to send text messages, for example. There one of those was lying around our office, and and and some litigation always got incredibly excited about that. So uh and you will know Gwillam and you know, excitement for patent attorneys can be a bit thin on the thin on the ground.

Entering AI And Founding A Startup

SPEAKER_04 15:22

No, I think I mean, yeah, well, I've definitely been involved in litigation. And every time I've done that, I thought, gosh, this is a lot lot more rapid than the uh day to day. But no, definitely exciting. I get nervous because you you you get all the decision time, are we gonna win or lose? And you get you get really caught up in all. Yeah, so I mean shifting to AI. So as you as you said, kind of in the last year or so you've moved to this kind of AI startup. Tell us a little bit about that, please.

SPEAKER_01 15:51

Yeah, so um I as you know, Gwillam, I'm very I'm I'm I'm really quite old, and uh I was uh I stepped up stepped down. Well, I'm older than you are, so you know I um I actually had stepped down uh off the board of Page White and Farrer. I was a senior partner when I when I stepped down in June 24, and I was on a two-day-a-week consultancy with them with a view to moving into, you know, like a proper sort of retirement with things like, oh, I don't know, you know, sailing and music and uh seeing friends and having time for family and all of that good stuff. I'd recruited 15 year years previously a colleague, very, very talented computer scientist. And for the last couple of years, before I left, actually, he'd been showing me various attempts he'd had at using AI in the world of patents. And and then in December 24, he, which was while I was on my part-time contract, he said, I've done something that's really quite amazing with the new reasoning models that have just come out. Do you want to have a look? So I did want to have a look, uh, and it was pretty amazing what the reasoning models could do and what he was harnessing them to do. In February 25, only this year, he said, I think that we can actually deliver this to the profession. I mean, gen AI is going to come into the profession, and uh, so far I haven't seen anything that really works. I can make something that actually works. Do you want to come along for the ride and and help me do something with this with this product and be the next wave of of bringing Gen AI into the pan profession? And I said yes, and then and then went and had a few more conversations with all the people I'd promised retirement to and said that this was a a new a new venture for me, a new direction, and and that's been it. So the company was found Atane IP was founded in uh in May, and we've been working hard to get the product to an MVB stage and deliver it to deliver it to customers. And the idea is that it will solve for many of the problems that are holding AI adoption back in the patent profession. And that's the piece I'm really yeah, genuinely uh inspired by. It's engineered to avoid hallucinations, it doesn't hallucinate. Uh in 2,000 cases, it hasn't hallucinated. You can see the reasoning, you can see how it's thinking, you don't have to just believe in its output. I mean, I could go on forever, Gwillem. Where you know, where do you want to take this?

Where AI Helps Most In Patents

SPEAKER_04 18:35

Well, I'm interested. I mean, yeah, we've got as I say, we on the podcast and in everyday life, you've got AI happening all the time in the IP world. Lee, you and I we did some, we did an interview with a business a couple of years ago when we were entering Atlanta, didn't we? You know, there's so many different offerings out there. Where do you see the niche? Where do you think AI is going to be gonna really break through with the patent profession?

SPEAKER_01 18:57

Well, that's a fascinating question. And that is a very, very good question because we're seeing attempts to introduce AI across the whole product suite of what anything that patent attorneys might do, all the way from portfolio evaluation, freedom to operate, identifying infringements, patent searching. I think that the where AI has an absolute sweet spot is in deep analysis of large quantities of material. That's where you give the humans a massive leg up. So you and I could read two or three prior art documents in, you know, a couple of hours, a couple of hours or something, compare them with a patent to see whether the claim's novel or not. But we couldn't do it for 50,000 documents in any in any rational time frame at all. And so I think where where it's being scaled up to do big projects, it's going to have real, real teeth. I think at the moment it's a little bit in its infancy in terms of things like freedom to operate, where it's sort of searching around for in potential infringements, comparing them with portfolios. But I think the next phase of that, where you're going to get some deep forensic analysis of what the AI has found in searching, I think that's, and that's really where our tool comes in as well, is to provide the next level of accurate analysis of documents and propositions rather than just a kind of broad AI summary type thing, which is where most of the offerings at the moment are at that kind of broad AI summarization type level. With some very nice, I have to say, user interfaces with dashboards and all and all things like that, probability indicators, uh scoring, which which is fine for for big uh, you know, fine for getting a feel maybe for what's going on, but it's not fine when you're at the patent attorney or patent litigator sharp end and you actually need to know the answer and you need to know the answer quickly.

SPEAKER_04 21:08

How how fast do you see uptake coming? People are suspicious of AI. You mentioned hallucinations. Um, there's a whole bunch of concerns that people have. Do you think mindset's going to change, or is it going to take a while for it to permeate generally, not just your product?

Adoption, Trust, And Global Differences

SPEAKER_01 21:24

That's a really interesting question. And the answer to that would have been different even earlier this year. So in spring of this year, I can say that there was a huge wave of excitement in the profession about adopting these tools, particularly in the US, actually. Probably the US is first to move, which won't surprise anybody. And uh a real thought that these could be great. Now in November, that first wave of excitement has definitely passed on because of these issues like hallucination, oversummarization, and the whole black box. Concept. You put something in, you get something out, you have no idea whether it's wrong or right. So patent attorneys, which is the people I'm talking to, are now becoming a bit more measured and a bit more cautious in their approach. I don't think adoption has necessarily slowed down, but I think firms are going to be more careful about their evaluation processes and trying to make sure they get a tool which actually is helping them. From a job, a geography point of view, it's also interesting. So US first, definitely. The UK is starting to become very interested. Germany is also moving at a reasonable pace into evaluation of Gen AI tools. By contrast, Asia is still taking the view that they don't really want Gen AI near their patents. And the Chinese patent office has actually said you can't put you can't file a patent application with any AI generated content, which is an interesting approach. And I've no idea how they're policing it.

SPEAKER_04 23:06

Oh, okay. Two questions on that. I've not heard that. Question one, policing. Thank you. No idea. I get that. No idea. Question two, does that does that domestically not doesn't affect me personally, but is it that domestically generated, or if is it going to be PCT global filing from elsewhere?

SPEAKER_01 23:22

Can't can't can't can't answer at the moment. I'm throwing it out as an interesting fact, which is on my list of things to chase down.

SPEAKER_04 23:29

Yeah. Oh, I mean, I'll I'll about to see a significant swathe of um patterns that don't work in China if we're not, which is a bit worrying.

SPEAKER_01 23:38

Well exactly, and that and that was that was what came out of the room. The room was quite shocked to hear this actually from the Chinese pand attorney who who gave this talk at AIPLA.

SPEAKER_04 23:48

Okay, thank you. That's a nice little nugget. Can you get nuggets on radars? Just open questions. But you know what I mean. I've got one more Leon, so I'm I've got one more kind of good one here, and then you're welcome to jump in. I think it's a good one.

SPEAKER_05 24:04

I'm very I'm very happy for you to carry on in the chair, Gwillam. You're doing a grand job.

SPEAKER_04 24:08

Okay, okay. So you mentioned before, Virginia, that there's AI tools out there for starting to replicate everything pattern attorneys do. And it's occurred to me that whilst that's an obvious thing to do in one sense, it reminds me a bit about that saying by Henry Ford when he said, you know, if you ask people do they want cars, they'll say no, they want horses that go faster. And so I think there's a risk that a lot of the time, rather than working at what AI could do that could change things, they're looking at what do we do that we can use AI to do instead. It's kind of wrong question today.

Beyond Faster Horses: Redesigning IP Work

SPEAKER_01 24:45

No, you're absolutely you're absolutely spot on. And the more I look at the capabilities of AI and the world of intellectual property and patents, and think how would this look with a clean sheet of paper, it's a very, very interesting thought experiment because at the moment a lot of tasks are separated out because of because you need humans, and um, humans, as I've already pointed out, are not good at like mass evaluation of anything. So you can't do tons of stuff in your head, and a lot of the tasks have grown up on that basis. So I mean there's a how I see it is there's a phase now, maybe five, seven, ten, I don't know, years where actually the products that work will will be those that do the patent attorney tasks as well as or better than a padd attorney in terms of accuracy. And I can tell you there's there's there's vanishingly few of those. At the moment, there's uh there's us. I mean, we can assume there will be others, but they're not there yet. Attain IP can is is verifiable and at the moment at least completely accurate. But I think down the line, there's going to come swathes of development through the whole profession. So we've already heard a big, big consultancy firm say what one of the things they'd like to do is set up an AI-powered IP practice. So from the ground up, what can AI do? What can that deliver into the IP market? It honestly, it's completely fascinating, but but it's a really, really excellent question. And and and have me back on three years and see what's see what's going on in the email.

SPEAKER_04 26:37

No, I think it fascinates for me. I think our thought processes are fettered by what we've been trained is possible to do. That's a sentence. And it kind of reminds me, there's a famous quote the reason we don't look at file history, this is very patenty and geeky now. But the reason is that we did look at file histories for years when we're doing litigation was because one of the judges possibly all this said, I don't want to look at a life file history when assessing a patent. Life is too short. And I think, in a sense, that's often what's in the back of our head. We have it's just not feasible to spend this much time doing it, but life is no longer too short to do some things. And I think we have to take us take a deep breath and say, Why aren't we doing this? Is it because it used to be impossible? And I think there's a whole bunch of possible stuff that we've not really actually had to look at now that could change a huge amount of how the IP infrastructure works.

Training The Next Generation With AI

SPEAKER_01 27:26

Yeah, and I think that's absolutely right. And you can take it even one step further because at the moment you have human v human patent attorneys versus patent offices who are kind of wrestling, if you like, to get patents out of the system. Now, if you take the view that AI can do a good accurate analysis of a patent claim, you can both mutually agree the novelty point using AI before you get up in get up and have your first cup of tea in the morning. Yeah, you'd be arguing about different different premises, you know. So I think it it almost turns into a philosophical debate about what does it mean for the premise of patents, which is that governments can grant these monopolies to companies and people based on a government examination of what of whether the monopoly is is is validly sought or not.

SPEAKER_03 28:29

Terminator timely.

SPEAKER_01 28:31

Very much so.

SPEAKER_05 28:32

Have you got me? Um we'll edit this bit out, but my Wi-Fi has been really, really flaky. I've been dropping in and dropping out, and um so uh actually we'll keep we'll keep it in because this is the great challenge with technology, isn't it? That we're um that we lock our we lock ourselves into it more and more so, and then we're at the whim of it actually working when you need it to. So um I guess that's a universal truth for us these days, isn't it?

SPEAKER_01 28:55

Well, well it well it well it is, and and and actually I don't know why it prompted this thought, but it's an important thought, which is one of the things we haven't touched on yet, but it it it's close to my heart is what does all this mean for the future of the profession and for young people coming in and attempting to train to be patent attorneys? Because if you take the view that within a few short years, tools are going to be available which can support senior patent attorneys in a very effective way, and I think that that that's gonna be the reality, actually. That does that mean that you don't need to train patent attorneys, and and of course it doesn't mean that because you still need to have people coming through who understand. I mean, they might be having a different understanding about how they work, but they need to understand what the core of a patent is and what the purpose of a patent is and what a patent claim is, etc. etc. So they need to be trained, and and so one of the things that we've built into our platform is is the ability to because it is a shows the reasoning of the AI before it delivers an output, it can be used to train because it's delivering additional information for a potential candidate.

SPEAKER_05 30:12

I don't want to take us down a particular rabbit hole because I'm conscious, Willem, that we're near time and I'm gonna need to move towards a closer. But it I I have a similar worry, but about assessment, Virginia, because I'm I'm conscious that the way we assess currently is always by looking to the past. So it's kind of like how how how did we examine last year? What should examinations look like this year? And we're not looking to the future and thinking what what skills and behaviours and competences do we need in the in the future profession? And if it feels like AI will move on a pace and the examination system, as in student examination system, not patent examination system, won't keep pace.

SPEAKER_01 30:46

Well, that's also yeah, that's also a that's also a thought. Yeah, I mean, and and professional development is an area that I'm looking I'm looking closely at. Um with AIPF actually, are you aware of Association of Intellectual Property Firms? Nice, nice group for IP boutiques, actually. And uh because IP boutiques are often the ones that do a lot of training into the profession, so they're very focused on these issues of professional development. So again, I'll have more for you in you know, another six months or a year as as these tracks start to move.

SPEAKER_04 31:21

I think I thought I thought I knew every acronym with a letter's IP, and it turns out there's another one.

SPEAKER_01 31:25

So I've given you two facts today, Grillem. Two facts that you're interested in. Two facts.

Desert Island Books And Closing

SPEAKER_05 31:33

I'm gonna try and squeeze another fact out of uh out of Gwillam, Virginia, and also out of you. So so the way the way the way we usually close the podcast is by uh coming up, or generally it's my job, sometimes Grillam surprises me and has one up his sleeve, but I try and come up with some kind of tangential question. So it needs to relate to what we've talked about, but it it's also kind of quite a whimsical way for me to have a little go at Gillam. First off, and only because I was listening to Desert Island Discs this week, uh I listened to it in the gym and was listening to the Tim Berners-Lee recording that was released recently. You you were you came into the profession at that rich time uh in the kind of four or five years before Berners-Lee created the World Wide Web and then went on to make it as influential as it was. Did you ever meet we asking? I did not. I was asking Virginia that, sorry. Yeah, I don't know.

SPEAKER_01 32:18

Sorry, sorry, no, no, I didn't know, no, sorry, no, no, no, no, I didn't know and I never I never mess him, yeah.

SPEAKER_05 32:23

That does that that does that doesn't matter. It would have made the question far more interesting if you had. Grillem, have you ever met Tim Berners-Lee?

SPEAKER_04 32:29

No, I have not met Tim Berners-Lee.

SPEAKER_05 32:31

So if you if you if you're not a great Desert Island discs listener, listen to his um Desert Island discs. It was absolutely brilliant.

SPEAKER_01 32:37

Uh I can only imagine, yeah, yeah.

SPEAKER_05 32:39

Thoroughly engaging. But I'm not going to ask you a horrible techie question, Gillam. Uh it's just my inspiration for it. So um, so I was thinking I was thinking about you. When I was on the treadmill running, I was thinking, what would Gillam do if he were on a desert island uh and someone had presented him with a copy of Shakespeare and a copy of the Bible, uh, or any other religious text of his choice? What other what other book would you want with you?

SPEAKER_04 33:01

Well, I was gonna tell you what I'd do with the Bible and Shakespeare, actually. Which is I'll answer a different question. I think I'd ask for a pair of scissors and some glue and see whether I can make any other books out of the words I cut out of the Bible and Shakespeare. That'd be years.

SPEAKER_01 33:21

I think I think you'd find, I think you'd find you could probably make virtually all books. Uh but I did, but I did hear uh or read somewhere, and I can't quite remember the percentage, but I think it was something like 80 or 85% of all of the like stock phrases that we use, you know, like those um you know comparison type phrases. I can't think of the word now what they're uh some someone help me out here with the word, these standards cliche type phrases that we use. Idioms. Sorry? Idioms? Idioms would do fine, yeah. 85% of them are extracted either from the Bible or from Shakespeare, and I think it's over 60% from the Bible, actually. Okay, okay.

SPEAKER_05 34:05

I I heard something similar recently. Yeah, I heard something very similar recently.

SPEAKER_01 34:09

Yeah. So you so I'm pretty that's why I'm confident, William, you go a long way with with the Bible and and and Shakespeare.

SPEAKER_04 34:15

I think I spent I spent my first year then trying to think of idioms that weren't either and then checking.

SPEAKER_05 34:23

Well, well that that that would keep you busy. Bearing in mind that you've got no Google to do that for or Chat GPT or other or other AI interface to do that for you. Uh okay. Well, I'm going to abandon my question then because that was far far more interesting than the question I asked. Uh, and just thank Virginia for coming and sharing her story with us. It's been for me, it's been absolutely fascinating to listen to that and um to get a perspective on what AI means for the future of the profession, which I think is probably our most taxing question at the moment. Gwenham, thank you for being a lovely co-host again, and I look forward to seeing you on the next one. And also my normal closer, and that's of course that if you've listened to the podcast, why haven't you listened to the podcast if you've not listened to it? If you've listened to the podcast and you found it really, really interesting, leave us a little review or something like that so that other people can find the podcast. Thanks.