[00:02]
What we’re really doing is providing space and opportunity and a better use of time for our clients to bring the greater portion of the workflow process to them. We’re doing what we can to help clients embrace technology and this transformative time and do it in a way that builds confidence and has this kind of overlaying security and protection around it. And if we take away the effort from the mundane, we’re freeing them up for the greater insights they can bring to our clients.

[00:39]
When it comes to navigating the future, Lumanity is devoted to helping its clients deliver on the promise of innovative medicines for patients around the world. Today I’ll be joined by two executives from Lumanity, Chief Client Officer Michael Parisi and Chief Technology Officer Anthony Guethert, who will delve into the evolution of their organization and how they are approaching the transformative impact of AI for their clients in today’s current environment. And this conversation will impact what surprised them most on their AI journey and what they are most excited about moving forward.

[01:06]
I’m Jack O’Brien, managing editor of MM+M and this is the MM+M Agency 100 Strategies for Success podcast with Lumanity. Michael and Anthony, it’s great to have you on the show here. Michael, I’m going to start off with you. Talk a little bit about some of the biggest challenges you faced last year in the context of new and emerging challenges as opposed to continuing trends that we’ve seen in years past. Well, Great to be here, Jack. Thank you for having us and thank you for the question.

[01:33]
Yes, so 2024 a really interesting interesting time for many of us. It’s kind of like the tele-two cities, if you were to think about it as a a headline in that there’s so much excitement in terms of scientific innovation, what we talk to our clients about in terms of differentiating their assets, their companies, their approach to marketing. All of that, I would say, was kind of the excitement piece. But then we’re faced with this very uncertain external environment.

[02:00]
I think one that we have faced before, but never as complex and all kind of consuming. And so you have this kind of convergence of excitement of technology and innovation that’s being met with this external environment of uncertainty. And that’s causing a lot of stress in how people are thinking about commercializing, bringing products to market, new approaches to market. And so it’s an interesting but exciting time.

[02:28]
And I’d say the rate and pace that we’re working at. The eyeballs on the business from our clients’ perspective and the need to really try and manage this uncertain time has added a lot of intensity into into the work. So it’s it’s really trying to align with client priorities, client needs at a time where things are just moving very, very quickly. It’s tricky.

[02:51]
It’s a tricky market, and it’s not one that anyone could say we have the answer to, but certainly we have some approaches to it. And one in particular as an organization, Lumenity is built around deep expertise in a number of areas. But it’s the nimble application and the flexibility of applying those different skill sets at different times. That makes it really important. So it’s one thing to have the skills, but it’s another thing to be able to apply those and tap into those skill sets as they’re needed in the market.

[03:20]
Yeah, you talk about being a tricky time. That’s something that I hear a lot from leaders across the industry. And that’s kind of where I want to bring Amy into the equation. It’s like one of the trickiest aspects is being able to keep up with the rate of change from a technological perspective and all the innovations out there, certainly the gen AI of it all. What has what posed the biggest challenge to you last year, Anthony, from your perspective? Thanks. And nice to be here with you today.

[03:44]
Challenge was really keeping pace, as Michael mentioned, not only with the rate of the marketplace and uncertainty, we’ve also got regulation change across other parts of the world hitting AI. The speed and pace of AI itself changes and impacting on our business is huge.

[04:03]
Then you’ve got governance being brought in over the top, then you have uncertainty in the marketplace, and bringing those three things into Coalesce and actually drive the business forward constructively is actually a real challenge. So what we’ve tried to do is meet the challenge head-on, find spots where the AI can be harnessed well, and where we can have some control points and yield some value from it.

[04:32]
And therefore, we can try and gain some certainty just in those pockets and areas where we can make use of it, but it’s certainly a challenge. And the pace of change and the improvement in the technology has been phenomenal. The one other area that has certainly created a bit of a challenge is just the share of volume and type of hype around it. Technology does go through hype cycles, and this is one of the largest I’ve seen. And with it, you sometimes expect there to be a bit of a crash on the other side.

[05:01]
But instead, what we are seeing is greater specialization and interest in it, which is, again, increasing in speed in certain pockets of our business where we can apply it. And again, that’s a challenge to keep track of, but also exciting opportunity as well. Definitely. And I appreciate both of your responses. Talk to me a little bit about how prepared you think your clients are for these transformations because there is all this sort of idea of like being able to stock up for when the moment happens. Are clients there? Still room for improvement?

[05:31]
Where are things stand? Yeah. Thank you, Jack. Great great question. And I’d say not just the AI component, but the rate and pace of change is definitely putting pressure on our clients. to think about and to do things differently. And I think depending on who you’re talking to, there is a tremendously wide range of adoption of AI in certain parts of clients and client businesses.

[05:56]
I think overall there’s an ambition, but like any industry there’s concern about security issues, privacy issues, especially our business where you have data privacy needs and they change globally and they’re changing by the minute. So really being ahead of compliance, both from a trial perspective, but all the way through to to how we talk about products and innovation. So it’s a it’s a wide range, and I’d say the industry is doing the best it can to keep up, but boy, is the technology moving quickly.

[06:27]
And so it does cause a little bit of pressure. We do spend actually quite a bit of time with our clients workshopping. We’ve released thought leadership pieces around things like the use of AI and publications. So very, very specific use cases. And opportunities. So, we’re doing what we can to help clients a, embrace technology and this transformative time and be doing a way that builds confidence and has this kind of overlaying security and protection around it.

[06:57]
Michael, I completely agree. This is very much a transformative time. I think it’s a great way of summing it up. The range of responses, as you’ve said as well, is vast and can make your head spin at times. We’re really seeing some clients embrace technology deeply. Like ourselves building in and creating platforms for use.

[07:17]
We’ve seen some clients back away and being rather concerned about the use of technology such as AI in this space, wanting to see how it plays out first and take more cautious approach and with some people, some of our clients who are in between seeking third-party expertise to help guide them as well. And we hope we can help there too. So it’s certainly transformative. We do think it provides though opportunity for us to show that expertise and help our clients even more.

[07:46]
And on the topic of AI, I’m really curious in terms of where that innovation is having the most impact on the business and considering that you both come from different sides of the business with client focus and technology, what is it meant for each of your lines? Maybe Mike, if you want to start off. Yeah.

[08:03]
Just would say one thing at Luminati that we’ve done and I’m really proud of what the way we’ve done it is while I come at it from a client perspective, I work very closely with Anthony and his team, and we really think about our business holistically and our clients’ needs holistically. So the way we’ve approached it, I think, is unique internally.

[08:22]
But if you look across, and I’ll ask Anthony to talk about very specific use cases, there are tremendous areas of efficiency and areas where we could really power up our insights like our market research capabilities, our market access capabilities, our communication and omnichannel offerings. There’s a high-level a number of different businesses that are embracing different types of technology.

[08:48]
And Anthony, I’ll hand over to you to talk a little bit more specifically about how we’re using technology in the business. Thanks, Michael. It’s been a fascinating year and a half, two years of journey. And Michael, your guidance and leadership on this has been phenomenal and bringing us together and working in tandem with our clients and the marketplace.

[09:13]
What’s really been essential is being able to be very specific on where the technology is used. The more specific we are, the more we can bring accuracy and our scientific strength to an activity, a subtask, or to a topic, or to a challenge or problem. And so by that, what we mean is we can address those concerns of privacy or hallucinations quite well.

[09:40]
For example, in certain areas where we’ve been looking at using market research data with generative AI to bring in synthetic personas and a product we’ve put together called emulator. It essentially emulates or models patients’ responses that we can glean from market research data. And we can normalize and segment that information together, those responses together and essentially create a synthetic persona for each segment type with patient.

[10:10]
That can then be given to medical science liaison teams, they can be given to sales teams, and the interaction can happen in natural language with a login into a browser into our platform. To allow these teams to to interact with a synthetic patient safely, quickly on a range of topics.

[10:31]
And we’ve worked in a huge amount of detail on the privacy and the accuracy removing the hallucinations to bring these personas to life and to have some real value and meaning back through to our clients. And it’s been fascinating working through that. It’s taken a huge amount of effort though to do it and to have the building blocks in place bottom up.

[10:51]
So we can address hallucinations and privacy as being a huge amount of work, but it does start to pay off when you see great results and we’ve seen solutions we can bring to our clients, which is really delighting them.

[11:04]
If anything, we’re working hard to keep the pace of the rollout of these technologies and check so that we can ensure that the value and the business and the onboarding and the training processes are in pace, lock step with the use of the technology as well. But it’s been a fascinating journey. And we’ve so much more to do together. So it’s been absolutely brilliant. Yeah, you talk about it being this I think the comparison I hear a lot is drinking from a firehose.

[11:34]
Like it’s just there’s always some new technology, some new tweak to existing technology that’s out there. And I’m curious how the agency has been able to tackle that and what is a rapidly evolving marketplace when it comes to the use of AI. Yeah. Anthony, I think it would be great to talk a little bit about Gale and and some of the fun we’ve had onboarding our staff and getting folks internally very comfortable, which has led to very specific use cases.

[12:01]
So Anthony, do you want to talk about the development of Gale and what that meant for us? Yeah, you’re right, Michael. It’s been a fascinating journey.

[12:09]
So quite some time ago with Michael’s direction, we saw the need for being able to use generative AI in a manner that was safe that allowed us to rapidly iterate, innovate, experiment, and then see where we could really lean into use cases that had value.

[12:34]
And so building a private platform, a platform where we’ve brought in privacy-by-design, the core principles of that privacy-by-design structure. We’ve built a platform on top of brought the market-leading large language models to.

[12:55]
and then applied very clear separation and segmentation into how each person’s session works, complete privacy is built into the each person’s session. We have folders and files you can bring to that platform, and you can use that platform in a number of ways depending on what you’d like to do.

[13:16]
And we started with a small group of subject matter experts who were tech-focused within the business as we started building out the the architecture of the platform. And by the time we got through to the first alpha version and went wider with the user acceptance testing and then went wider to the the whole organization our first release last year. Gale, a generative AI for Luminity platform.

[13:41]
The use we monitored closely began with just a few hundred people and then it grew and then it grew and then it grew. And organically within gosh, I think it was about six to eight weeks, we saw an uptake of greater than 70% of the organization using it on a daily basis. And then we started drilling into what were people using it for.

[14:03]
And the very specific sub tasks in their own specialty areas is where they started to use the platform most and some of the most creative ways that we when we’re building the platform had never envisaged we’re starting to be used. So in essence what our job became then was not just providing the platform and ensuring privacy and security work. are intact. And also that we’d respected our clients’ needs for conflict of interest and their use of AI.

[14:31]
So at the very beginning when you enter into the platform, we show you which clients are uncomfortable or do not want their data brought to it. We have very clear separation in place built into the platform. We also allow people to interact with the best practice prime and prompts which have been created by their peers. And we allow the use of those prime and prompts to be rapidly redeployed into a new use case as as well.

[14:55]
So, people can take existing content, they can take existing prime and prompts and make use of them really quickly. And what’s been really, really fascinating is people you’d least expect to embrace technology have become some of our biggest some supporters of the technology.

[15:14]
And if anything, we’re now finding that we’re needing to be very robust with clear business cases and return an investment framework to ensure that we really focus on the best examples or opportunities in a really structured manner because there are so many good ideas and opportunities we can apply technology to with you know tasks and activities that we’re doing for our clients.

[15:38]
It’s been a brilliant journey and the the level of embracing enthusiasm and the intelligent application by our subject matter experts has been phenomenal. And if anything, I feel a real sense of pride being in this company and seeing the extent of scientific rigor and expertise been brought and used so well with technology. It’s It’s been phenomenal. Yeah, no, it definitely comes through in droves here.

[16:03]
And I’m curious and maybe Anthony, you can tackle this one first in terms of maybe what The most surprising discovery has been for you on this AI journey because I’m sure you came in with some expectations of oh the technology could deliver this but what has anything surprised you? Has anything stand out? Yeah, a number of things have.

[16:19]
And I I think the key thing that really has surprised me has been the I suppose the creative way that people have found uses for it and embedded it into their work practices such that Gail has become a verb. Um you know you’ll be in the kitchen area, you know, making a coffee and people will talk about, “Oh yes, I’ve gailled it. I’ve done this, I’ve done that.” and they’ll be talking about an activity or a task.

[16:48]
And I suppose the biggest surprise to me is the fact that people have taken what we’ve provided and stayed within our clear governance guide rails, but then really leaned into it and applied it in such brilliant ways. Um, the ability ability, for example, to provide key talking points in a workshop, to be able to provide ways to communicate with each other.

[17:16]
So when you’re in a technology-driven team and you’re thinking about how to have a difficult conversation on a technology topic and how to distill it down and communicate it well, oh well, you know, I put it through Gail. I did this, I did that, and it came up with these suggestions on how to communicate this across to our clients, and this is what and it’s phenomenal. And you look at it, and you think, actually, that’s really good.

[17:38]
People say, “Well, I could have spent 2 hours doing this.” However, it took me about 3 minutes and it’s come up with something better than I could have created in 2 hours. You think, “Well, yeah, that kind of makes sense.” But then when people also start to bring scientific publications into the mix and what we have also implemented is retrieval augmented generation rag. And what that does is it essentially allows you to upload a scientific publication, for example. It’ll index it into memory of your session.

[18:07]
And then Then, when you start querying the content of that publication, it will actually show you where in the publication is deriving answers from. And what that means is our subject matter experts are able to be very, very specific and precise on answers that were deriving quite quickly from scientific publications by referring exactly where in that publication and answers come from, and then they’re able to bring in other publications to try and cross-reference, validate, or contrast.

[18:37]
And And just watching that evolve and then providing the right tools for them has been incredible. And I think one of the biggest surprises was when we built an extension of Gale to work on systematic literature reviews at scale and speed. Having the engine build out 22,000 responses across 100 publications in 2 hours was incredible. And I think that was one of my a heart moments.

[19:02]
Seeing that level of responses back and then going through sitting with two PhDs and looking at the accuracy level was also really surprising at how we can hone and improve and tweak the accuracy even further up, which has been great. So we’ve had several surprises. It’s been absolutely brilliant. I know there’s a lot more to come. No, definitely. And the world of AI doesn’t stop and between the time that we record this podcast and put it out, I’m sure there will be a dozen new entries out to the world.

[19:32]
So that’ll be something for our audience to keep an eye on. If if there’s one other thing that I’d also mention it’s that while the technology has increased and brought efficiencies to some of the more mundane tasks or some of the more data intensive tasks, the teams have turned around and used that additional time to create far deeper insights and far sharper analyses into the data itself.

[19:59]
So what we’re really doing is providing space and opportunity and a better use of time for our clients to bring the greater portion of the workflow process to them. It’s the insights that our subject matter experts bring to a topic that is the value. And if we take away the effort from the mundane, we’re freeing them up for the greater insights they can bring to our clients. And that’s probably one of the other big value points that we’re trying to surface.

[20:29]
And Anthony, I’ll just I’ll add to that. All well said. I think when we started this journey, Anthony and I were spearheading a small team to really embrace AI. And what we thought at the time would be there would be pockets in the organization that are ready. You know, you always move to your first first your fast adopters.

[20:49]
And And what we’re really proud of is how we were able to roll this out, reduce the fear and get people comfortable by essentially playing around with Gale and learning different applications. And from there, developing sub-teams that could get very specific back to your original question, Jack, around the different types of service offerings and how best AI plus a human will increase the quality of the deliverable to our clients.

[21:20]
and I’ll steal a quote from one of our brilliant scientists, Tim Becker, and he said, “My gosh, it just helps me get off the blank page.” You know, I start with the blank page, and that’s where I struggle, and this gets me off the blank page quickly, and enables me to bring people around and start to solve problems the way I normally do. And so we’ve had tremendous uptake. The creativity surprised me.

[21:43]
I really am blown away at how different people are bringing use cases forward, and thinking about how to, to Anthony’s point, make the mundane a little simpler, so we can take the brilliance of the individuals in the organization and make sure that comes out in everything that we do and deliver. And I’d say we’ve gone from 0 to 100 miles an hour very, very quickly in terms of uptake. Again, I really appreciate you both being on the show here.

[22:10]
I want to conclude, obviously focusing a lot on everything that happened in 2024 for the agency. Talk to me a little bit about what excites you for 2025, what’s coming up on the horizon? Well, Anthony has a long list of stuff he and the AI our tech team are building. So we’ve got a lot of stuff we’re excited about. I mean as an organization that’s now we’re still young as an organization and there’s a lot of awareness of Lumanity, but not a lot of deep knowledge about all the value that we can bring to clients.

[22:40]
And so we’re excited to continue to foster deeper relationships with our clients, share new things and new ways that we’re able to help solve their problems. But I’ll go back back to the biggest the beginning of, we’re living in a time of uncertainty and helping our clients figure out some best practices and ways forward in this uncertain time is some of what we’re excited about. Scientific innovation continues to drive our industry.

[23:08]
We’re excited about the future and some of the impact these therapies will have, but we’re equally excited about how we bring those messages forward, how we engage patients, patient communities and clinical trials. trial development.

[23:22]
I’m a creative person at heart, and I’d say this is an incredibly rich period of time for us to think creatively about the future and think about what we could do differently to help accelerate access to medical advances and really bring these therapies to people faster. So it’s exciting time for sure. Anthony, I don’t know if you want to to add to that comment at all. Yeah, thanks Michael. I completely agree with all of that.

[23:46]
And as a technology-focused person too, I think the team and I are really interested in the greater biotech and biopharma specific technologies that are coming in as well. So, for example, some of the small language models and some of the large language models that have been specifically trained on really good curated data in our life sciences space show real promise.

[24:14]
Some of the small language models we’ve built with curated data have yielded really good results and have been wonderful at sort of improving the way we can use AI for some of our own use cases and persona development, for example.

[24:32]
So I’m really excited to see that continue and seeing other third-party tools becoming more available to help people get off the page, as you mentioned, Michael, and also improve the quality of what we can communicate across to our clients and to our patients is one of the key reasons we’re here. So we’ve got a lot more to come in 2025 moving into 2026.

[24:56]
And I think there’s a lot of very good tools that we can apply to improve what we’re doing even further, which is brilliant. So thank you. It’s um Yeah, it’s been amazing to see so far. It’ll be amazing to see what else is around the corner for us. Definitely. No, we’re going to keep focused on it in the future and thank you again for coming on here to talk about all stuff of Humanity and and congrats again on being listed on the Agency 100 this year. Certainly well deserved. Thank you. Thank you, Jack. And open invitation for you and the team. Come visit us in our Morris Town Hub office.

[25:25]
We’re not that far from from your office. We’ve moved everybody in about a year ago. It’s a great space and we’ll look forward to hosting you. Absolutely. Once we get this New Jersey transit strike fixed, we’re going to be right over there. Wonderful. Excellent. Thank you again. Thanks. Thanks so much, Jack.