Richard Nicholas: Legal Foundations for AI in Private Equity
Summary In this episode of the AI Pathfinder for Private Equity podcast, host Steve Budd talks to technology lawyer Richard Nicholas to explore the intersection of AI, legal governance, and risk management in private equity. They discuss the challenges and opportunities that AI presents, particularly in terms of legal compliance, data privacy, and the importance of effective governance structures. Richard shares insights on how private equity firms can navigate the complexities of AI implemen...
Summary
In this episode of the AI Pathfinder for Private Equity podcast, host Steve Budd talks to technology lawyer Richard Nicholas to explore the intersection of AI, legal governance, and risk management in private equity. They discuss the challenges and opportunities that AI presents, particularly in terms of legal compliance, data privacy, and the importance of effective governance structures. Richard shares insights on how private equity firms can navigate the complexities of AI implementation, emphasising the need for collaboration, communication, and a proactive approach to risk management. The conversation highlights the critical role of in-house lawyers and the necessity of fostering a culture of trust and innovation within organisations as they adapt to the evolving landscape of AI.
Takeaways
- AI projects often stall due to legal and contractual issues
- In-house lawyers are crucial for managing AI-related risks
- AI governance requires collaboration across various business functions
- Trust in AI systems is vital for successful implementation
- Policies alone are insufficient; practical governance is necessary
- Engaging employees in AI processes enhances adoption and effectiveness
- Innovation and governance can coexist to drive value
- Identifying potential risks is crucial for AI deployment
- Continuous learning and adaptation are key in the AI landscape
Chapters
00:00 Introduction to AI in Private Equity
02:41 Richard Nicholas: Background and Expertise
05:04 Legal Challenges in AI Implementation
07:48 Managing Change and Resistance to AI
10:20 Blockers to AI Adoption
12:42 Ongoing Governance and Responsibility in AI
15:22 Legal Foundations for AI Implementation
17:39 Six Steps for Effective AI Governance
20:09 Building Trust and Managing Risks in AI
22:38 The Role of Private Equity in AI Governance
25:17 Conclusion and Future Insights
Links
Richard's LinkedIn: https://www.linkedin.com/in/richardnicholas/
Browne Jacobson: https://www.brownejacobson.com/
AI Pathfinder: https://www.aipathfinder.co
Private Equity AI Strategy Day: https://www.aipathfinder.co/ai-strategy-day
Private Equity AI Strategy Breakfast Briefings: https://www.aipathfinder.co/breakfast-briefings
Have a question or would like to suggest improvements, please contact steve@aipathfinder.co
AI Pathfinder Private Equity AI Strategy Meet-ups:
https://www.aipathfinder.co/private-equity-ai-events
Have a question or would like to suggest improvements, please contact steve@aipathfinder.co
Hello and welcome to the AI Pathfinder for Private Equity podcast. I'm on a mission to help private equity firms go from AI curiosity to confident action by building a network that uncovers what's working, where the risks are, and how to make smarter decisions when it comes to AI across the dual life cycle from origination to value creation. In that mission, I get to meet some brilliant people who are tackling AI from different angles. And today's conversation is going to fit firmly in the legal governance and risk category. While the tech might be impressive, it's often the legal contractual data issues that cause AI projects to stall, or worse, create unexpected risk after deployment. Joining me today is someone who knows how to navigate this. And if I quote his LinkedIn profile, known for giving clear answers in short sentences, we'll put that to the test today. We're going to unpack the legal building box as much as we can in the 30 minutes we have for successful AI mobilization and how private equity can help their portfolio companies get this right from day one. Richard, welcome to the podcast. Thanks so much for joining me. Thank you. Good to see you. Yeah, and you too. Well, let let's let's start with you, Richard. Could you provide a little bit of background on yourself, the sort of work that you do, and what's driven you to specialise in this area of tech and AI in particular?
SPEAKER_00Yes, thanks, Steve. So uh I've I'm a technology lawyer, have been since well for the last 25 years, since so since 2000, working on technology projects. So lots of IT procurement, software development, licensing, that sort of thing for a very long time. Mainly for large businesses, retailers, banks, that sort of thing. And for the last, well, certainly since about 2016, I've been uh quite heavily involved in data and GDPR when that came out, and obviously before it came out, we got uh just over a hundred uh clients through their GDPR journey, got them compliant with um GDPR, got uh their governance process in place, and have for the last couple of years been working on AI, which really brings together sort of technology law, uh, procurement, uh contracts issues, and also the governance structure of GDPR. So it really combines those two those two areas. So that's uh that's where I work. A lot of my clients are a lot of my clients are in-house lawyers at these sorts of uh at these sorts of businesses, so I do lots of uh training in particular for the in-house lawyer or the DPO or whoever manages risks in the organization, mainly with a view to uh sort of managing risk. And and why this area, Richard, what what what makes you sort of gravitate to tech and AI? Well, besides it being sort of the the things combining things that I've looked at previously, so combining IT and um and data, it is I find a fascinating area to look at, partly because as a as a lawyer, it's not that often that things come up that are genuinely new. This is one of those where actually the the problem solving involved, it brings together data, it brings together IT, it brings together IP issues. And so for some of the some of the clients that we're working with, we're looking at things like what does consent mean if you're if you're looking at data, and how can you use consent if you're assigning IP rights because you're you're um putting IP into a a database that's then going to be used for machine learning, and how on earth would anyone get their, how on earth would anyone exercise their their rights to see that information or to change that information once once it's into the s into that system. So all of those sorts of issues are interesting issues, and it's an area where everyone is finding their way. But we're now at a stage, I think, where there have been some ground rules set, and it's it's very much an international area as well. And had the privilege of being at a a large conference j just a week ago where some of the largest businesses were talking quite openly about how they're dealing with some of these issues and the sorts of sort of governance issues that um that they're seeing and and what they're putting in place.
SPEAKER_01Okay, well, look, I I think that it's we we know each other, we've known each other because we're curious about this emerging tech. And I suppose we've adapted to some change and maybe we we we look for change and we sort of head head um head straight forward into it. But yes, um others are a little bit more fearful. And I wonder not not necessarily going right back to when ChatGPT came came out, but uh to the public. But it'd be interesting to know how legal conversations have changed over the say 12 to lots 12 to 18 months. What's keeping the boards and CEOs up at night? And and yeah, so so be interesting to see we've gone this big learning curve, haven't we, to now where you know there is we're going from the education and proof of concepts to to production. So it'd be interesting just to get a bit of a feel for what you're hearing.
SPEAKER_00Yes, certainly. Looking at the there's a side of the board, if you like, that I see, which is often the chief legal officer or the head of legal or general counsel. And it's it's interesting. The sorts of questions the sorts of questions I get from them are first of all, how do I there's two two big questions. First of all, how do I do my job? How do I make sure that I'm managing risk with this new technology? Where do I where do I even start in terms of getting the right governance in place? What does that what does that mean? But also, and this happens sometimes in the sessions that I give to those lawyers, they'll sometimes come up to me afterwards, and there's a second question they'll ask, which is okay, but how does this affect me? How does this affect my own role? And what what should I personally be doing? And what are you seeing from other in-house lawyers and what are they doing? And that's sort of there's a there's a bit of business as usual protecting the company, but there's also a oh actually what does this what does this mean for me going forward and what does this what should I be learning in order to best be best adapt to AI? Um I think lawyers are a conservative lot by nature. There are still some lawyers and still some law firms who are saying we are banning it entirely. We're not going to even consider the i idea of AI. There are still some who are saying we don't want to be first. Well, I think that trains I think that trains left. I think that would have been the case back in 2023. There are some there are some others who are pushing forwards and actually innovating, finding new ways of working. And I think the the conversation is slightly changing now, partly because people are realizing this isn't this isn't a hype thing, this isn't going away. This is something that everyone is going to have to sort of uh adapt adapt to. And I think we're all going through the I'm I forget if it's the the grieving cycle or Alcoholics Anonymous, I forget. It's either, you know, the sort of going through denial, then embarrassment, then negotiating with it about, you know, which bits, um, which bits of legal practice we're prepared to concede to AI. And finally eatance and working out, okay, this is this is a tool, it's going to be a tool that in some ways competes with parts of a of a role of a lawyer, and how do we best best adapt it and use it to get the best results from it? And we're seeing people at different stages of that process, and I think that's true of actually everyone in the business, and some of the some of the best I've worked on various AI projects now, some of which have worked out well, some of which less so. The better ones have been the ones where everyone gets involved, so the people actually doing the job actually get involved right from the start. The less successful ones are those where someone in senior management has an idea to adopt AI, imposes it on the rest of the business, and inevitably, because there will be problems with it, those people who feel put out by the imposition of AI or feel threatened by it, they're the ones who will find those problems, will raise those issues, and that's often when the the project doesn't work. So I think there's a real there's there's some besides the legal and practical and commercial sides of AI, there's actually some there's actually some sort of people management as well in terms of actually introducing AI in a way that is takes into account what people might be feeling about the the process and how they might feel about uh accepting it and how you actually um upskill those people and uh empower them to use AI to do their jobs better rather than um them seeing it as a threat to what they've been doing.
SPEAKER_01Yeah, I I've um uh yesterday I was at an event for data and information providers, and obviously there's a they have awesome IP because they've spent a lot of money in uh curating this sort of data set that their customers need, but both from an internal perspective of how their teams are using that data um and maybe feeding it into a an AI that's not quite authorized, and but on the flip side, um you've got a customer that wants to do additional um work on the data and and and it's left the building, you know, you don't know quite what's happened with it. So there's uh uh obviously alarm bells ringing, them trying to ring fence this this data, but at the same time um trying to uh uh adapt with the new technology, it can be very, very difficult. It sort of brings us on to my next question, really, is is what you're seeing as the biggest blockers, if you like, for for companies trying to implement AI. And um and that can be you know internal processes, productivity, efficiencies, or it can be maybe more around something that's that's customer-facing.
SPEAKER_00Yes, I think there's a there's a few things. One of one of which um I mentioned already, which is that actually how you how you get people to accept it, buy into it, and feel empowered by it rather than threatened by it. And there needs to be a process, part of that is actually just sensible process management, getting getting the right um plans in place and getting uh getting communication right at the start. Um there are some challenges. So there are one of the surprising things I found about um AI, because I've now been negotiating with lots of sort of AI providers, and one of the things I found is having worked in IT for a long time and worked in software and SaaS and that sort of thing, it is quite difficult to negotiate with some of the very large suppliers at the terms of a SaaS license, for instance. Very difficult to change how that software is going to be provided to the business. Um and usually that's because their same software provider will be providing to lots of different companies, they will have set their terms and unsurprisingly they're not going to interchange them. Well, AI contracts are not like that. The space is not as well developed. Some of the contracts are some of the contracts are not good enough, so there are some there are some real issues in terms of ownership of IP, ownership of IP outputs and and the inputs and the prompts and those sorts of things. Um and the ongoing the ongoing monitoring. And actually what I found is some of those contracts can be can be negotiated. They are much more fluid. So there is there is more of a process in terms of onboarding um providers of AI. I think I think companies that go through the process really need to work out the problem they're solving and making sure that the AI fits. And that can make it a longer process, so that can be a that can be a more of a challenge. There's the transparency element of working out exactly what the thing does, and someone within the business, so again, unlike unlike traditional software where someone would just um buy the software, carry out acceptance tests, and then that's that's almost the job done. The software does the work and it's um it's accepted, signed off, and it's it's off someone's list. AI solutions, because they continuously evolve, they're continuously learning, and the outputs change over time, what that means is you actually need someone to take responsibility for it. You need someone to take responsibility that's close enough to the tech, close enough to the service line to understand what the AI is producing, but also someone who is responsible overall, who can make the decisions when when there has been a breach, when they need to switch the thing off, when they need to actually uh take some other sort of evasive action because the the the AI isn't producing the results it previously was. And I saw a um there was a an MIT study that looked into the risks of AI, and it came out with something like 350 of them. So it's one of those one of those things not to read if you've just if you've just implemented an AI solution in business. But just over half of those were actually risks that would crop up not at the implementation stage, but sometime la sometime afterwards. So there are various things that you need to keep on top of, and that's that I think makes a difference. And it's something that, again, it it needs to be part of that process. So you need to have decided at the beginning when we take this on, who's going to who's going to continuously take responsibility for it.
SPEAKER_01So you have um you mentioned GDPR earlier, so you had you know you had someone responsible, I suppose, to that. Is it something similar that you're describing here?
SPEAKER_00It is. No, it it it is. It is the regulation of AI is very similar in many ways to the regulation of GDPR. So you need a person who's responsible, and ideally, that's not it's not actually one person, but it's a committee. I speak to in-house lawyers all the time, and they're I'm I'm saying to them, you really do want a you do want a committee, you want someone in sales and in marketing and in finance involved in the process, partly because if you don't do that, it's likely to be you, the in-house lawyer, who has to make the decisions and who has to deal with the with breaches and deal with problems and deal with issues when they when they come up. Um but also you want actively want stakeholders involved in the sort of implementation of AI, in the same way that for GDPR, you need parts of the business involved. They need to know what they need to do. So it's it's it is a bit like um everyone in the business knowing how they deal with customer data. That's the only way that you get the business to actually be compliant. You need the same with AI. You need all everyone in the business to understand what AI can do, the limits of it, what they should and shouldn't be doing with it, because otherwise, similar to GDPR, um having a policy won't be enough. So having the the regulator, and actually it's the same in the UK at least, it's a lot of the work that's been done in terms of regulating AI is the it's the same regulator, it's the ICO, so the the same regulator as for data. Um, the regulator is not interested in a set of words on a page. So similar to GDPR, where one of the frustrations would be often you have clients who have just taken a policy from somewhere else, put it on their website, and they think that's the job done because it's they've got the right words in place. It's the same with AI. The words don't matter. You can get an AI policy quite widely and from a range of different places. If you're not doing the right things, then actually what's said in the policy won't help. So again, it's about changing practice. And the other way it's similar to similar to GDPR actually, GDPR changed the rules on data because what it did was it made it made businesses have a positive obligation, a bit like health and safety, they couldn't just not do the wrong things, they actually had to start doing the right things and showing their decision making. So if you're relying on legitimate interest for instance, you need to do a legitimate interest assessment, that needs to be set out. That doesn't go anywhere, it's something that you store, but you store the decision so that if the regulator comes comes knocking, asks why this happened, you can show them the note that was made at the time of why you made that decision. Well, it's very similar with AI in that you need to have those decisions recorded at the same time. They don't necessarily go anywhere. There's not a place for them to be sent. But if there's an audit as to how the company got into this mess, why the AI has been making decisions that are biased, which are discriminatory, which are offensive, whatever the case may be, you need to say, okay, well, what was the decision making at the time? How did you come to the decision that this was a good idea? And if you can't show that, you can't show that there was a positive decision made, then you are likely to be in more trouble with the regulator because you need to the the decision making is part of the governance process.
SPEAKER_01Yeah, yeah, okay. So you've given some good guidance there in terms of well, like we we are seeing committees, but often it's committees to um look at the the technology, not look at the necessarily the governance perspective. But um I think that's uh you know an im really important point for for my audience is to to understand where that sort of responsibility lies, and uh it's really important to have this sort of audit trial in place. So uh adding to that, what other sort of legal foundations should a I mean it it could be any company, but an SIP back company should be thinking about. They could be at the start or very early in their their process that there's still time for them to put these foundations in place. What what what do you think those two or three things are? I can I can give you six if you like. Six? Absolutely.
SPEAKER_00This is quite um it's very similar to the sort of uh in-house lawyer question, really, about you know what what do I need to do. I mean one of the things is is to first first, just in case people are in this mindset, because this has been promoted on the media as a as a battle between innovation and governance, as if those two things are opposed. Partly it's getting out of that mindset because actually you you can you can innovate better and you will get more value from that innovation if that is properly governed. It doesn't mean going slower, it doesn't mean holding things back. Often you can move just as fast. Often it's just a question of you know taking taking the notes as you do it, showing what you're doing and why you're doing it at the same time. Um but there's really six um a six-stage process which I've been running through with clients for some some time now. And having been at the conference and hearing from the likes of sort of MasterCard and AstraZeneca, I was relieved to hear that they have similar processes and similar frameworks in place that aren't that that different. There's a an AI mnemonic that goes with it, which is penguins can glide through rocky caves. And they're really there you go. Six um six things to do, and they're actually paired together. So the first is the P and the C, that's the planning and communications. So the first thing is to actually work out what you're going to use the AI for, make sure that there's a problem to be solved, making sure that you're not just buying an AI because it's a bright, shiny object, which actually does happen more often than you would think, um, and and getting your communications right, so it's both internal and external. Internally so that you get the employees on board and making sure that people buy into the AI and make the best use of it, but also externally so that customers know what you're doing and appreciate the value that you're going to get from the AI. Um that's the first thing. So planning and communication, there's governance and training. So the governance is having the governance structure, the sort of thing that we talked about about having the right people involved, having a having a process that will be triggered when things go wrong, a bit like a sort of data breach. So making sure there's the right the right people there, making sure that you're keeping on track of the results from the AI, and training to make sure that that isn't is throughout the organization so that people know that there needs to be a level of AI literacy. Now that's that's a requirement in the EU under the EU AI Act, but actually it's it should happen here as well, it should happen anywhere where AI is involved, and so that people know what they can and can't do and the limits of AI, and if they do innovate, how they how they best do that, because there are ways you can encourage people to actually use the AI to create value, and you want to be doing that. You want to you want to be create a an environment that encourages that. And finally, risk and controls. So there's a there's almost a brainstorming exercise required in a business to work out all what could what could possibly go wrong with the AI. And you won't capture everything, but that should be tailored to the business. So there are different types of business dealing with different levels of personal data, dealing with some places where AI may not be the most appropriate solution, in some cases a human empathy is required by customers, and working out working out that those risks and keeping a log of those risks and controls that you've put in place to manage them, and that might be having a human in the loop, so a human actually overseeing the final decision making process, it might be having a human who's responsible. If you've got planning and communications set up and governance and training set up within the organization, and you've been through that risk and controls process, you've worked out what the risks are and how you might mitigate them, that puts you in a much better position when it comes to either a regulator or If they come knocking. And typically that would happen where something has gone has gone badly wrong. It's all about and it also gives both the employees and customers a level of trust. And actually that's the that's the big thing with A in particular, is that there is a you need it's only as good as people have continue to trust it. And if you lose trust of customers because of an incident or because the um because the AI starts to produce results that it wasn't previously, or even if there's a perception that that's happened, that's when where you get the real risk, is a sort of reputational risk to the business. And if the business is built on AI, if it's a if it becomes an AI first company, then this is this is really quite significant for the business. It can make the difference between customers choosing you and choosing someone else. So having that framework in place, I would suggest is is the basics behind good governance and making sure that you're on top of the issues.
SPEAKER_01And it's not just a case of putting a one-liner that says the results might not be accurate.
SPEAKER_00Yes. This may not constitute legal advice. I don't think it is. No, that's right. So it's having having that's one thing, but then you know there are so many. There's um he might have seen the I think it's the website, is it incident incident.ai, of all the sort of AI-related incidents that you've seen. And some of those are not things, some of those are are places where the AI has w has done its job, it's done exactly what it was meant to do, but the outcome wasn't expected. Um, you know, there's the there's the Amazon, you know, the simplest form of this, the Amazon, you bought this, you might like this um algorithm, which um gave some particularly unpleasant results when when a book that was that had suicide as its as its subject um was uh was promoted. It promoted um other items and and things from Amazon that uh you would not want to be promoting uh in that situation. There was some th and again the AI was doing exactly what it was meant to, it was finding things that were related to the product that it was that it was related to, and you wouldn't necessarily have thought of that in advance, which is why you need that sort of ongoing governance, because some of the things that will go wrong with AI, you don't know yet, because you can't you can't possibly think of everything in advance. So that's where that's where sort of having that ongoing process is is needed, I think.
SPEAKER_01Yeah, and I think that as as businesses become sort of more aware of where the advantages are, is perhaps looking at narrower use cases so they're able to perhaps make or have better controls over the over the inputs and outputs because it's in such a sort of narrow, narrow space. Or or using AI, small language models where there's potentially less risk of hallucinations. But but as you say, it's it's a constant sort of uh review process. Um and it's something I want to sort of from from my audience, you know, the the the private equity audience are looking at themselves and and how they um they manage a lot of sensitive documentation, but they're also seeing that there are advantages of using using AI to um you know for a number of benefits, not just in terms of efficiencies. Um but there are also um custodians over their portfolios and which there can be several. Do you think that there's an opportunity, or or do you think it should be the PE firm themselves that sort of set those standards? Or do you see other ways that that there could be some advantage for a PE firm to to perhaps utilize the the knowledge around that you know that that portfolio and and within the the firm? There's too potentially too many questions I'm gonna ask here. And you were talking about you know this this ownership. Should that be perhaps there there should be a capability within the PE firm that can then sort of support the group, or should that be delegated across the portfolio, which could be cross-sector, um, cross-region?
SPEAKER_00I think I think you probably want some of some of both. If you're looking to build value in the individual companies in a portfolio, then you want those individual companies to have their own governance process, I would suggest, because you want ultimately, if it can when it comes to a sale, you're going to want to make sure that that company is standalone, that it can support itself and it has its own has its own process. But I think a layer on top of that is that every PE business ought to have their own governance and to work out what they are work out what they are prepared to accept and to to not accept within it within their own own business as well.
SPEAKER_01Do you think that that can become part of their sort of investment thesis, their their differentiation and the way that they conduct themselves in or how they um well it's that that equation you talked about, wasn't it? Sort of innovation, the governance.
SPEAKER_00I think so. I think at the moment it's it's a bit like the Apple taking the privacy as its as its sort of differentiator. Um I think it is worth doing there's a there's some marketing value in getting governance right and in showing that you're getting governance right, and is in many ways it shouldn't be a differentiator. This is what everyone should be doing. But I think if you can do that and you can show the value behind it and you can show the the value creation, I suspect I suspect we're not far off a number of different incidents that will affect companies that have have brought in AI and are not regulating it. At that point, I think people will legitimately and genuinely be asking questions, what was what was behind this? What have you got in place? I think we'll see it much more in due diligence exercises for company sales. I think it will come up in um I think it's the sort of thing that we'll see reported in earnings um reports and those sorts of things. I think we'll we'll see what is happening with AI and what level of um innovation a business has been through in those sorts of places, which I think will I think will inevitably make it much more of a public issue.
SPEAKER_01Yeah. Um sort of ethos really of what what I'm doing with um with the network is to try and bring examples, um make it accessible um in all the different sort of areas that they need to try and get their head around. Um and I just wonder, so if we're coming near to the end there, obviously I don't want you to share a sensitive information, but is there something that you've been involved with recently that's taught you sort of something new around this topic and that would be useful for for the group?
SPEAKER_00There have been I mean the besides the besides the project, the the getting everyone involved that does come from a live from a live project where it was a the most highly paid person in the room who decided that was something they were going to push ahead with. I think the sorts of issues I I think one of the things that that comes to mind is actually getting people together to work on this when where you've got people learning from the ground up. I think one of the one of some of the best examples of projects I've seen has been where individuals have been given the ability to use AI and experiment with it, and actually to learn a bit about how it works and why it works. And I think it's I think it's actually incumbent on most of us to learn how it works and how they can be how you can best use it because that gives you a that gives you a sort of degree of um self degree of agency compared to buying products that promise the earth, some of which actually can be some of which actually can be recreated within a business by enthusiastic individuals with who are who have been encouraged to innovate, people find the best solution. If you've got a a lot of talented people who are good at their job, getting those people to use AI is almost, almost inevitably going to be better than buying an AI solution that does what they do. Because they will know what's good, they'll know whether the the result is any good, they'll know the sort of subtleties to look out for and the things that can go wrong and the sorts of issues that customers need or want, or the questions that customers don't always ask, but they actually need. And so if you've got talented people, I would suggest getting those people acquainted with and up to speed with AI in such a way that they can feel that this is adding to their job adding to their role. That makes any implementation so much easier, and it also has the benefit of actually getting better results from what I've what I've seen as well, because the the people who are the people who are encouraged and actually actively interested in using AI in their role, they're the ones that make a a a positive difference. They're the ones who actually find the find the benefits, they'll help the implementation, they'll they'll get over the obstacles, and they're there almost inevitably will be, they will make a much smoother transition. I'd suggest getting some some champions in each area where you want to implement AI, getting people who are who are genuinely keen. That's where I've seen the best, the best sort of smoothest implementation, if you like.
SPEAKER_01Oh, great advice. And it you know, it comes back to people, doesn't it? Again, so it's change management and and those people that are are super close to that uh use case are absolutely the right people. It's I suppose it's a blend of personalities, really, about you know how that you've got someone that's curious, you've got someone that's that's maybe quite analytical, um and that they can they can experiment because you know we have uh for very little money tools that are very powerful. And so obviously we want to be using them safely in a governed way, but it presents so much opportunity, as you said, versus going up there and thinking that just buying a tool is gonna be the answer. It's probably not the immediate answer, that's for sure. I think that's a great place for us uh to leave the leave the pod, um, which has been fascinating and gone very, very quickly, too quickly. But you're joining us, aren't you, for our breakfast uh briefings that we hold quarterly. So you'll be there to answer questions, I'm sure, of those in the network. But leads me to say thanks so much for your clarity, brevity, penguin, penguins, lots of things for the listeners to uh go away and think about what the practical steps are. And look, there's in the show notes there'll be Richard's details of his LinkedIn profile. So I uh I would suggest you connect, and also details on the on the briefings that I mentioned. But for now, thank you very much to Richard and thank you for for listening. See you next time. Thank you.