MLV100 2026 Launch Party (AI Edition): Introducing the MLVx AI Platform

Harding's gone.

elsewhere. You know me, I'm Sean, and my co-host today is Jose, but we also have a panel. MedTech Leading Voices. So we are going… this is the launch party, woohoo! As you know, so for the past several years, I've compiled a list. The first year was 25 people worth following on LinkedIn. Why'd I do it? Because I couldn't find… on LinkedIn who to follow. It doesn't help you find MedTech experts. Like, go ahead and try, you know? So it became a manual process of building the list. And from that, MedTech Leading Voice was launched, which was, you know, 10 or 12

Posts per week that are worth reading. You know, and it became a newsletter. And from that, we started the webinars, and the webinar series, became the foundation for our body of expert content, which we're gonna explore today. But before I get into that, I want to say you know, my relationship with AI goes back exactly 3 years. 3 years ago this week, I had brain surgery, and it went very, very badly. It was called guided ultrasound, and I'm not gonna name the device manufacturer, or the hospital, but I'm gonna say that, I was… supposed to walk out of that surgery. It was, you know, non-invasive surgery. They didn't cut my head open, they just shaved it. And,

I… I was a mess for 6 months. I was an absolute mess. And I lost all manual dexterity. I couldn't type. I could barely walk. I had so much edema in my brain from this surgery that I could barely function. And then they put me on a super high dose of steroids for 3 months, which made me insane on top of everything. More insane than I usually am. So… What happened was, since I couldn't type, since I couldn't take notes, since I couldn't barely function, I started recording everything. And from the recordings, I started transcribing everything. And from the transcriptions, I started using ChatGPT

to analyze the transcriptions and to work from that. And what I discovered in doing this is, why did I ever bother taking notes in the first place? I just record everything, you know, and I'm able to do things with Claude and with ChatGPT that I was never able to do before, and basically have been able to do the, 10 times the amount of work That I was doing on my own, manually, slow way, writing articles, blah blah blah, all that stuff that I did, all the marketing, everything. So, as Jose says, you need to learn how to do 50 hours worth of work in 3 hours. And,

And that's the point of all this. So… We started to develop this body of content, and all of these experts, and all of these newsletters, and all of these slide decks, and all of these videos, and all of these transcripts from all of you, and from all of these meetings, and we decided to compile it into a single searchable AI-driven tool, which Jose is going to introduce today, called MLVX.ai, and you can log into it right now, and it's free. To use, and many of you are in there, already, and, And you can ask questions, and so I'm gonna turn it over to Jose to bring it up and walk us through this amazing thing that he built.

And if I haven't thanked him enough, I'm doing it now. Thank you. Enough.

Awesome. Thank you. Yeah, no, this has been a pretty fun project. As Sean said, I think, you know, we're at the advent, really, of a crazy era that's coming upon us. AI is incredibly powerful, but at the same time, we all know that AI is not perfect. Humans aren't perfect, AI is not perfect. Hopefully, together, we're closer to perfect. And so this, this platform has started brainstorming around it. There's a couple elements to it. So, the first element is that… the thing I've always loved about this call, going back even, you know, years to when Joe, headed it up, is that it's not like a one-way communication, it's really a community, right? So, there's… there's conversations. I mean, I've come to know many of you through here, and then the conversation continues on LinkedIn, and then we've had Zoom calls, and

And so, you know, now I know that when I get a client who needs help with 6601, I know Leo, and if somebody needs help with QMS, I know Nellie, and somebody needs with regulatory, I know Carrie, and so it's like, I've got now a community of experts whose judgment I trust. and whom I can refer to my clients, or whom I can reach out to myself if I've got questions. But of course, it's not always practical to reach out to everybody at the same time, so what we're trying to do is bring all that knowledge base and all that expertise into a place where people who have questions can get better answers that are more grounded in expertise from people like you.

Or if you're on that side, then you can get that. And at the same time, as Sean mentioned, there's not a really good place to find experts in these various fields within MedTech, so we're also trying to do that. And so, for kind of, like, consumers of this knowledge, you know, medical device startups, medical device OEMs, it's a place where they can get better answers and exposure to experts in the field. And then, of course, for the experts, it's a place to really, get exposure, you know, to present yourself and to show your skills and show your knowledge base to other people. So that's the vision behind it. I'll say that this is, like, early days. We haven't been working on this for too long, so…

It's… we decided not to wait until it was fully polished to start showing it to you guys. This is, like, a soft launch in our little community of people whom we know and love, so that you can give us feedback. And that feedback can be anything from, like, hey, Jose, there's a bug, or hey, Jose, or Sean, like, I really wish we could also do X, Y, or Z. And I'll show you in a little bit how you can do that. But that's… that's the grand vision behind this. So without further ado, I'll give you a little bit of a walkthrough. When you, when you come to the website, if you haven't signed up, you know, you're still able to, like, see, the various experts. You know, a good chunk of these are MLV100, and I know that's another purpose for this call, is to present to you the MLV100 2026.

And so there's only a little bit over 100 experts on the platform right now, because what we did was we brought in that full list of the MLV100 as the starting group of Experts to be on the platform. And then, of course, over time, as people, want to be showcased as experts, they can sign up and, you know, start providing content. So that's the first piece, and then there's an Ask AI feature, which is, you know, like your standard ChatGPT, or Claude, or Gemini, except that it's grounded on the information that comes from experts. So I'm gonna go ahead and click to sign up. There's two ways to sign up right now. You can sign up through your LinkedIn, it's our preferred path, you know, if you're open to it, or you can do it through, a Google account.

I'm gonna do through Google, because I'm gonna sign in as, like, not my main admin account, but as, like, a secondary account. And on here, what you'll find is, you know, kind of like your standard, again, chat. Thing, except it's a little bit different in that you can see on the references where it's pulling information from. So, for this particular question, I asked this question because, you know, I write a lot about cybersecurity, and so I knew I'd come up here. So what cybersecurity documents do I need to submit to FDA? And what it's going to pull from is we have over 100 FDA guidance documents, we have dozens of guidance documents from other geographies as well.

We have those transcripts from all those webinars that Sean has hosted over the years. We have blog posts that experts have, you know, right now it's primarily mine, because you guys haven't come on here to link to your blog post yet, but when you do that, your blog post will be part of what feeds this. And so the response is going to be based on that corpus of well-vetted knowledge. And it's gonna be… you know, cited to the experts themselves. So when I asked this question, you know, it's primarily drawing from this first reference, which is the 2026 FDA guidance on cybersecurity, but, you know, it's also pulling some information from me, like, I did this, you know, blog post of, like, five

reasons why FDA will refuse your 510K, you know, or cybersecurity requirements, or, like, all these various things. And on any of these, I can always… I can see who the experts are who are cited here, and I can use this to go to their profile page, where I can also, click on this, and it'll go to that source. So I'll go ahead and try that real quick. And so you see what it does is it goes to my website with the blog post that I have on there. So, as an expert, one of the reasons why this is good for you is because this is now bringing people who come to ask a question to your website. And that has a couple benefits. One of them is that, of course, now they're kind of more in your sphere and can see what else you're up to. Number two, it's good for search engine optimizations, because now you're getting backlinks to your website.

So that's… that's the benefit to the experts. And… here, let's see… If I go back here, and then on the expert side of things, I can click on… on, Jose's… Paige, right now, we're pulling a lot of this information right from your LinkedIn profile. So, you know, if you went on my LinkedIn page, it looks very much like this. And… and as experts, when you have an expert account, you'll be able to add your own call to actions, your own CTA buttons. So right now, as an expert, I set up this one, which is, like, our Cybersecurity Quick Start Guide, which is a document you can get on our website. But then also, you know, you can schedule a meeting with me, that comes up.

And then, you know, you can ask about me, that kind of stuff. It tells you… it's like a shortened version of your LinkedIn profile. actually noticing that my LinkedIn button isn't on there, so I'll make sure that's added. But, like, if we go to Sean's, you'll see that I can click on this and connect to his LinkedIn profile. So that, you know, I can connect with them on LinkedIn, which is great. Okay, what else can I show you?

So, question.

Maybe before I go much farther, let me see if there's any questions.

I have a question.

Yeah.

So, I'm a little slow when it comes to these types of tools, getting on board on them, so my question is, how easy does, like, for example, do I post here and then automatically post to LinkedIn, or do I post to LinkedIn and it links back? I mean, what's the direction of communication?

So, from a posting standpoint, we actually haven't added that as a feature. We've got some ideas around that, and, so kind of like a live feed of short posts, that's not something we've added. It would be more like if you're, you know, here, I'll go back to my profile here. On here, you know, you'll be able to post things like content, like blog posts, like longer form stuff. You'll be able to post, media, like videos and slides. And then, we'll be able to pull in like, webinar transcripts and things like that. So, right now, it's meant to be more, like, long-form content that's being pulled in here.

Library, like a library, not an instant post.

Correct, yeah. And so… and, you know, that's one of the things, like, here, you're gonna see there's… there's now a… pretty good, video library, so there's, like, 63 different videos on here. I'm not logged in as an expert right now, so that's one of the differences. So here, let me… let me log out. And I'm gonna sign back in now with my LinkedIn. account. Okay, cool. So now, when I go, or is it? Okay, yeah. So now, because I'm logged in as an expert now, I can add… I can add videos, right? So I can say, hey, here's, like, a thing, either through URL, or I can, or the videos are all through URL. If… the other option that I have is to upload slide decks.

And then those, you can do a URL, or you can upload a PDF. And when you upload a PDF, it actually… it runs it through another large language model that takes a crack at, grabbing the title and a brief description of it. So, like, for example, if we looked at this slide deck. From Basel, you see that this enhancing risk management through The product lifecycle, this was… pulled directly. And then, like, a brief explanation of it. And it's like, we're just getting started, right? Like, we're still coming up with ideas. That's part of, like, the role that you guys play, is to share ideas, like, hey, Jose, this is cool, but it would be even better if, let's say, you could link this to the webinar.

Or, you know, if I go here and I look at a video, right now, we don't have the full… We don't have the full transcript located here. Even though the transcript of this has probably already been ingested, we don't… we don't show it here yet, so over time, we'll add those features. And, and real, you know, a real easy way for you to do that is when you log in, if you just click here, there's gonna be this send feedback button. So if you click on that. you'll be able to, you know, is it general feedback, is it a feature request, or is it a bug report? And then you'll just tell us your subject and a description, and submit the feedback.

And we're gonna try to be really responsive, you know, obviously, if it's some gigantic feature, it'll take a little bit longer, but if it's like, hey, it'd be really helpful if you just added this small thing, I mean, maybe within a day, we're able to make those kinds of changes. And bugs, of course, like, if you find bugs. Please report that back. So those are… those are various ways. Let me… there's other stuff that I can show you, but before I do that, let me just pause. Any… Questions, comments…

Yeah, Jose, I'm logged in right now, and I was looking, and it says I'm not an expert, If you're an expert on MLV and believe this is an error, please contact support. I'm contacting support, I guess.

Yeah. Good, yes. So that's a great opportunity, Leo, to do that. And one of the things that we're gonna do as part of this, I know a lot of the people on this call are experts, and some are maybe more like, hey, we're just here to learn, but, everyone on this call, we've got… so there is… okay, there's a cost to this, long-term, right? Of course, there's a cost to us, like, LLMs charge, But, one of the things that we're doing is, if you help us spread the word, like, if you post about this on LinkedIn and just tag Sean or myself, you're gonna get 3 months for free. And it's, like, I think it's reasonably priced for what you're getting. If you come on here on the pricing page.

You can be on here for free, you know, the limitation is that You're only gonna get 5 questions per day if you don't register. If you do register, still for free, you'll get up to 10 questions per day. But it's not gonna save your, like, chat history and stuff like that. The pro version is $49 a month, you get everything free, you get, like, 30 questions a day now, chat history, etc. There's a Pro Plus version that gives you now 100 questions a day, and then there's an expert version, which, you know, again, like, if you're part of the expert category, and you just post about this on LinkedIn and copy us.

then you're gonna get 3 months… well, any of these, you'll get 3 months for free. So if you're not an expert, you just want a Pro Plus, you'll get that free for 3 months. like, right now, we're kind of in the early days, you know, for most people who want to be involved in this, we're going to give you guys a promo code so that you can do that, but over time, there will be that. Once you create your account as an expert, you're going to be able to link it to your… to your expert profile page. And then you'll be able to do things like, you know, for me. there's linked here, I can actually edit my profile. So if you're not registered as an expert, it hasn't created that link between you as an expert and your expert page. So, obviously, we don't want just anybody being able to edit somebody else's page. So, once you register, it sends us a message,

We confirm that you are, in fact, that person, and then we link them, and then you can come on here and do things like edit your profile, You know, title, company, etc. Your specializations, you can choose up to 4. We're trying to keep that fairly limited to… so that it's easier to search the specializations. You know, your credentials, you can add all that. Other stuff is, analytics, so you'll, you know, again, we're building this out, but you'll be able to see if you've got an expert account, like, how many profile views, how many people have come and looked at my profile, how many people have ended up clicking to come to my website.

You know, for your CTAs, how many of those have happened? You'll be able to upload your content here. So, videos, slide decks, articles, etc. Add CTAs, so these are the ones that I've added, so you can schedule a meeting with me, or the Cybersecurity Quick Start Guide. etc. So that would be what the… Expert page looks like. And, Yeah, and I'm sure there's gonna be kinks that we have to work out, so Leo, I'll take a look at what you got, and… And we'll do that.

So there were some questions about, different browsers. Somebody tried to do this with, I think, Safari, and they couldn't get in. Or…

Interesting. Alright, well…

and… So there was a question, it was, is there any restrictions, Safari, Chrome, Firefox?

There shouldn't be.

And then…

Yo.

Someone else… Asked, does this come with MLVX membership? Yes, for the time being. Anyone who is a founding member of, MLVX, you're… you're… you're golden for the rest of the year. The, so we can figure that out. Because we only had 38 people join as founding members, and I didn't really push that very hard. But yeah, you, you guys have, A huge advantage. And we'll get preferential.

treatment let's see, the link is not working, show… I tried to write, all of the, blurbs for the MLVX100, And basically, you know. what value you provide and why follow you, right? So, each one of those blurbs is written like that. They're being updated. If they're not up to date yet, we're… we're working on it. But the… yeah, so… There's an example there.

And.

Kept them pretty brief, right? But yeah, shares pragmatic FDA regulatory Strategy for Shell. We're also going to change it. One thing that right now is a restriction, and we're going to remove it, is on the, Like, if you don't register. Right now, you actually… you can see all the experts, but when you click on them. prompts you to register. We might change that so people can see it no matter what. But all you gotta do, you don't need a paid plan, all you have to do is just register for the free version, and you'll be able to see people's, profiles.

How are you gonna draw in… because really, the reason we're part of the community, one, is to build the community, obviously, but two, it's to find new clients, right? And so that's the reason… so how are you going to make this attractive? Because when you're looking at startups. They don't have… every penny counts, right? Every penny pinching, it counts. And so, what model are you going to use? I mean, is there going to be a different model for, hey, these are people who are advertising versus, hey, you're a startup, if you're under this cap, we're going to give you this rate so that you can come in and look at experts. What are you going to use to attract

Potential clients for the base. group.

Okay, I can, I can address that. First of all, it's always free. Like, you can… you and startups and individual users can always use this platform for free. Now, you're gonna hit a limit where it's gonna say, yeah, you've asked enough questions today, right?

Oh, okay.

But there is… there is… the free tier is… is free forever. Am I correct? Jose? Yep. Am I lying?

No, you're not lying.

Okay, so that's… that's really important to understand. There's… they have access to, all of the basic functions, just… they're gonna hit a limit on the number of questions that they can ask every day, because it costs us money. You know? We're… we're then paying for their chat GPT, or their Claude, you know? And that, we're not that generous. But the, So, yeah, that… that's the… the one thing. And then, The other thing is that an expert Can also be on here for free.

But… they're… they're vetted. We're not just, like, hey, you want a profile, is that correct? Or is…

I'm the experts, Well, right now, the linking of an expert account to an expert profile that's existing is… there's… yeah, there's, like, you… you request to be linked to it, and then an admin will review it, and then accept it, and then you'll… you'll have access to that page. But some of that stuff we're, honestly, we're, like, fleshing out right now, right? So we're starting with, hey, we loaded everybody up there, we created the accounts, you know, this is, like, very much a beta launch, and… We haven't worked through those little kinks of, like, hey, what happens if somebody wants to create a brand new expert account, or what happens if somebody wants to be linked to an existing one, or…

Also, like, what happens if you don't want to be listed on here? That's easy. Like, if anybody on here, they're like, hey, I don't want to be on this page, you just send us a message and we'll pull you down. So, I mean, we assumed most of the experts that are selected for the MLV100 are people who like to post. And, and so, presumably want exposure. But obviously we respect everybody's wishes if anybody didn't want to be on here anymore.

Oh, and the other thing is, is that the URL is just mlvx.ai. Right. That… just type that into your browser.

Nope. Yeah, I'm curious to see… Carrie, what is… what is… did you say yours didn't work? What… are you just typing MLVX?

There's a moment in time before coffee.

It's fine. Okay. Got me scared for a second there. Same thing with the Safari, I mean, I just checked on mine, and it works, loads up…

No, I think Sean… can I blame Sean? There was a… there was a… It's fine. I broke it.

Blame me, I don't care.

I don't know, I wrote the wrong… I didn't write .ai, I wrote .com. So don't write .com, write .ai, and then you're fine. I thought. Dunfu.com. then it won't work. If you do .i, then it works just fine.

Okay, great. Thank you.

Thank you so much.

clarifying that.

PublicServiceannouncement.ai.

Thanks, Carrie.

You're welcome. I'll go back on mute.

So, I wanna… I wanna circle back to this idea of ephemeral… ephemera, you know? Like, the content that's getting created on LinkedIn, the webinars that are happening, the events that are happening, they're… they're a point in time And it's like the news cycle. They fade within… within a week. So, the goal for me was… there's a lot of value when… when 2 or 3 experts get together and they discuss a topic. For example, earlier this week, Maria Donua and Danielle Giraud, did a fantastic talk on 18969 on clinical evidence, which is the new standard that's being developed for clinical evidence.

And… You know, nowhere else do you have access to a convener and a stakeholder of… of… The standard, that… come on and discuss what the standard is, why it exists, where it's going, what it's supposed to do, what it's not supposed to do, answer all the questions around it, and… and then it goes away. And it's like 460 people went to the event. But what about the rest of the world? And so, the idea was to take that content, to capture that content, and then to expand that content, so that you're able… to interact with it and use it to your own advantage. And so, for me, it's like, I have all these files, I've got the text files, I've got the JSON files, I've got these files, you know? And…

but they weren't helping any of you. And the same is true with LinkedIn, to a certain extent. Like, a lot of you, are doing really… Beautiful infographics. and, explanatory things that show up for, what, a day, two days on LinkedIn, and then they get buried in the sea of, you know, non-recoverable information, right? Marina is on this call. She produces just a tremendous amount of content, and having an archive of that content that is searchable, accessible, and attributable to her. On biocompatibility is going to be extremely valuable for, all of the search engines going forward who are relying on

authoritative content to understand complex issues, particularly in niche areas like medical devices, if we can call it a niche area. So, the… my point… around all this was, as an archivist, as a person who looks at this and says, whoa, whoa, whoa, why… why are we letting all this stuff go? And why are we at the whim of the search engines? Why are we at the whim of SEO? Why are we at the… whim of the algorithm, behind LinkedIn. And I don't want that, you know? Personally, I want to know who can I trust on this topic. And so that's, for me, the motivation going into this is, like, to say, no, no, no, we're not trusting you to give us the answers. We're gonna give us… we're gonna give our own answers, and we're gonna build a community.

One word that I'm starting… to both love and hate, because I hate jargony words of any kind, but ecosystem is to develop a medtech ecosystem, which is a community, which is content, which is resources, which is interactivity, and I really think that AI is changing the shape of knowledge in the sense that… I know that's a garbage marketing expression that I just used, sorry about that, but the, your ability to master a topic relatively quickly with AI, or to feel like you've mastered a topic relatively quickly with the help of AI,

is… is phenomenal, so how do we then ensure that it's actually factual, and that you can trace it back to the guidance document, you can trace it back to the expert, you can trace it back to… Nellie Bushman said… She said, this is what we need to do to get our QMS in shape. Carrie Britton said, this is what we need to do to prepare for a QSub meeting, you know? It's not this unattributable content that is just floating around, like, well, who said? Well, ChatGPT said. Okay. You know… Do you really want to go to your boss and say, how do you know? ChatGPT told me?

You know? I don't. I don't even like my paws.

Yeah, and just to add to that, I mean. you know, like, somebody asked, and one of the questions was, like, are you, like, vetting the experts as they come on? And I think at first. Probably no, right? And so, people need to understand that LLMs, like Large Language Models, ChatGPT, Claude, can make mistakes. They can hallucinate, they can confabulate, they can synthesize, sometimes they call it. So can humans, right? Humans also make mistakes. So, you're never going to get an answer that you know 100% certainly is true. And by the way, like, even when you read a guidance document, the guidance documents, those are, like, let's say, accurate, but

They have to be interpreted. So, you're still, as a human, have to use your intelligence and your judgment, where you're gonna say, well, look, I'm coming to MLVX because it's giving me the perspective of experts. Now, are they… you know, am I assured that every expert on here is 100% right all of the time? No, because they're human beings, and they might be misinterpreting something, but at least I'm getting that. And at least I'm going to a place where I can see the input from various experts. Over time, we do want to add additional features, so… so, for example. like, what it's doing right now, maybe I'll give you a little bit of a sneak peek into what's happening in the background.

In the background, what this is doing is this has something called, I'll get to you in a second, Carrie. It's got what's called a rag. I guess by the panelists, raise your hand if you've heard of a rag before. Maybe one or two, not too many. Yeah, so it's a technique that's used pretty regularly, so it's retrieval augmented generation. So, the way a large language model works is, you know, you feed it a prompt, and then it goes into, like, a statistical model, is really what it's doing. And then it's, like, literally token by token, which is oftentimes not even a full word, it's estimating, like, what should be the next word on this, right? So, we don't have to get into those details, but you're feeding it context, like your question, it's going into this model, it's spitting out a response.

The coloring, or, like, the details, or the quality of the response it's gonna give you. Is a function of what you're feeding into the model. And so that's going to be your prompt, but there's actually also what's called system prompts. So, meaning, like, Anthropic, the way you can't… the reason you can't ask Anthropic, like, how do I make a bomb, or something like that, is because Anthropic has system prompts that tell the model not to answer certain questions. Or to, like, you know, be careful not to say certain things. So there's… there's already another system prompt that's going into The model before it gives you a response.

The way that a RAG works is that it's got this database of information, and when you ask it a question, the first thing it does is it looks to see this question is there content in our database that is similar to this question, or that seems to be related? And it does that with, like, semantic kind of calculation. I won't get into the details of that, but it's just like looking to see, like, is there other content that we can load into the prompt. to give us a more grounded response. So what we're doing in this system is we're taking all of this content from experts, and we've built out this rag.

So that before it sends that prompt to, like, let's say ChatGPT… well, we're using primarily Anthropic, so Cloud 4.6 is the model that we're using right now. before it sends it, it's grabbing content from, like, you know, let's say from a guidance document. You ask a question about cybersecurity, and the first thing it does is says, okay, well, I've got a guidance document, cybersecurity, I've got the E-Star content, I've got this blog post that Jose wrote, I've got this, you know, webinar that, you know, Mark was in. And they all seem to talk about cybersecurity, so I'm going to bring in the relevant parts of those, those conversations or guidance documents, and I'm going to feed all of that with your question to the large language model, so that the response it gives me is

More grounded on that, rather than just its training. Does that make sense? So that's what's happening behind the scenes, and in fact, we have our own system prompt in this, which has certain guardrails, like. Don't answer based on your training, answer based on what's in the rag. So there's going to be certain questions that you ask today, where it's gonna say, like, I don't really have anything in my system related to this, so I can't answer that question, but maybe here's some, like, experts who… whom you might want to reach out to. And then, as we add more expert content, then it's going to be able to answer more and more questions, right? So it's… we're trying to be really careful about having it not

Hallucinate or just give made-up answers. With that, you know, there's always a trade-off. So, somebody in the chat had asked, like, hey, what if I already have a subscription to ChatGBT? It's like. I don't envision this replacing ChatGPT or Claude or Gemini. Like, those tools have their own purpose, and they're very powerful in different ways, they just have their own drawbacks. So, I can see myself coming on here to ask certain questions, and maybe even asking the same question on Claude or in ChatGPT. to get another point of view, because ChatGPT today will answer a question based on its training, based on… it'll do a web search, it'll do all this kind of stuff, so it's just getting information from wherever it wants.

And it's gonna be… and many times, it's a great answer, right? But sometimes it's not gonna be a great answer. So, you're gonna get different answers depending on whether you're asking MLVX, or whether you're asking just a standard chatbot model. But that's… I don't know, hopefully that gives you a little bit more sneak peek into what's happening, You know, behind the scenes on this. Any questions, or…

Yeah, can you show Maria, the… our references embedded in the information that you just described? How?

Can I just say, while he's doing that, I was just gonna say that with this panel of experts, what you're getting is a human in the loop who has that human experience in the loop. who knows those references and can say that, you know, FDA just released ESTAR 6.2, which now makes you now submit your bundle of QMSR SOPs with your submission for your next 510K, and I don't think ChatGPT may not know that, for example.

Yeah, that's right. That's right. Oftentimes, the training for these models are kind of at least a few months behind, and now they're pretty good about searching, like, they can do web searches and stuff like that now. But you also, you're not always certain, like, where I got that information. So this gives you… and what I love about this is just that it gives you access, right? So I had asked the question earlier, what can you tell me about ISO 18969, you know, which Sean mentioned earlier, and we've got Maria on the call here, and it's…

Seminar, by the way, Maria.

And it's pulling… it's pulling that information right from that discussion, and now it's… it's kind of enriching it with… with its own, judgment and stuff. But it's primarily drawing from it, you know, and it's like, right here, it even quotes, right? So, as Donova summarizes, it's supposed to make sense, it's supposed to help, etc, etc. So you're getting kind of that… the quotations from the documents. And then, if I want to view the sources, I can click right on here, and I can see the full transcript of that conversation. You know, we'll see if maybe there's ways that we can improve this down the line. And I know my back-to-chat right now is broken, so…

have to come here, but we'll fix that. That's an easy one. But yeah, it's gonna be pulling from that, and it's gonna give you, you know, access to, like, who those experts are. I've asked them many questions over the last couple days, obviously, but… yeah. So, like, this one, which is a cybersecurity question, it pulled content from me, because I write a lot about cybersecurity. I've already uploaded, like, more than 50 sources to this, I think. But it also pulled something from Angelina, and I was like, that's kind of interesting, because, like, Angelina writes more generally about, regulatory things.

And to my surprise, when I looked at it, it was actually a transcript from a webinar that Muhammad and I did right here. And, but Angelina, you know, said some stuff. And so, it captured that, and And so now I'm like, oh, look, Angelina, like, she's got this experience, and there's one video with her, and… And all of that.

I have a question here.

Sorry for not raising my hand. Oh, it's fine. Hop in.

Hi. Elon, my name, hi, Sean.

Hi.

How are you? Just a question that can be, regarding, privacy here. Are you somehow have access to who asks what, why, when? or this is complete compliance with, I don't know, CCPA or whatever.

That's a great question. So, so first off, we don't, we don't store, like, we don't have access, even the super admin on the account, which is me, doesn't have access to any of the questions, or responses given. You know, at best, what we get is information, like, if you click on a… even that, though, like, if you look in the expert, dashboards, so every expert account is gonna have a dashboard of, like, profile views, content views, CTA clicks, but it doesn't tell you who clicked on your profile. And that's a feature that some people might be… like, LinkedIn does that, right? If somebody comes to your profile, they can show you who did it.

we're kind of erring on the side of privacy, of saying, like, you know, I shouldn't be able to see your questions, because they could have proprietary information. Every model that we're using, you know, has a privacy, policy in place.

to the API, so the API is the most secure way to… yeah. Correct.

That's a great question. Yeah, I'm glad you asked that.

Thank you.

Yeah, I'm glad you asked it, too, I didn't know the answer.

Yeah. Alright, who else has questions? I see Daniel asking to be promoted.

I can do that.

Yeah, can you promote him?

I just did it.

Okay, cool. And then, Maria, you said, are references embedded in the information that you just described? Bye. Yes? Yeah, yeah, so the references would be… yeah, like, here it's telling you, pre-sub must clearly identify the purpose, and the blah blah blah, and if I go on the references, it's pulling primarily from this guidance document, which I can click to, and it would take me to the FDA, so, like, request for feedback. And meetings for medical device pre-submissions.

I wanted to ask Jose a question.

Yeah.

Let's assume for a moment, going back to the privacy discussion, that somebody actually, by accident, puts information about a product, or a process that is just, you know, that belongs to a specific company. So now, I go back into this… the question and searching process, and I ask a question about that company. It… that information is already in the system. Am I going to be able to capture it?

No, I mean, the system… the RAG on this system is strictly from sources that are uploaded by people, right? So it's… there's absolutely no training based on questions or answers in the chat.

I see.

Yeah, the only info… the only way to get information into that database that it's pulling from is, you know, as… like, as an expert, if I went in here and I went into the dashboard, and I said my content, and I uploaded, like, a new video, or I uploaded a slide deck, or I go into this knowledge base. and I can upload a PDF, or I can upload a, like, a URL to my blog post or something, then it'll pull that data. But it's never gonna store or capture the data that you put into the chats themselves, no.

And I think this is something good, obviously, because even though I wouldn't do it, but if somebody did it by accident, at least they will have the, the assurance that, you know, there's nothing that is being staved or stored. Okay, great. Thank you.

Great question. Thanks, Anna. Yeah, Michelle.

Yes, so…

neighbor.

Yeah, I just realized that about 2 weeks ago, so… My question is that while you're doing your demo, I see a lot of questions that, you know, like, on the left side of your screen. I just logged in, and obviously, I probably need to build my own questions, because right now, when I tried to click on the questions, they were just empty and nothing.

Well, you're talking about these questions here?

Yeah.

Well, these are my questions, right? So this is saving my previous searches, so you're not going to see those, because this is linked to my account, right? So to Daniel's question earlier, just like, I mean, ChatGBT and Claude do the same thing. Where if you've got an account, if you've got a paid account, then you can always go back and look at the history of your own questions. But I can't go look at the history of, like, everybody's questions, because, you know, again, that would kind of be a privacy issue. Yeah, but as you start asking questions, now, I think right now… I have to go back and look. I know right now we're saying that if you register but haven't paid.

If you're not registered, it's not gonna save your questions for sure, because it's like, how can it link it to you if there's no account? I think right now, even if you're registered, it doesn't save your chat history. Like, we gotta give you a reason to upgrade, because somebody's gotta pay the bills around here, right? We're trying to give you… give people a good amount of value for nothing, but, you know, we're hoping that people find enough value in this that they're willing to, you know, pay what I think is extremely reasonable. Even for the experts. I mean, I'm really excited about the expert, because, you know, it's like one of those things where it's like, I don't know if any of you are old enough to remember this, but there used to be a commercial for, like, hair growth that was like, I'm not only the hair club president, I'm also a client, right? Like,

that's how I feel about this, like, I'm not just excited to be building this with Sean and his team and my team. I'm excited about this as an expert who wants more clients also. And, you know, I have a… all kinds of ideas of how we make this even more valuable for the experts. Like, I think this can become, like, one of the… it's kind of a bigger feature, so it might take us a little while to build it out, but, you know, there's a lot of platforms to do courses. I think, like, wouldn't it be cool to have, like, a single place where it has all the courses on…

you know, these topics that are interesting to me, and I know, you know, places like Raps and other such places have these, too.

But I think… for… for experts who just are like, I just want to post my course on here, maybe it's a paid course. I think that's a feature that might be cool to build out. But I think… we want to get a lot of these ideas from you. So, don't be shy if you're using this, and you're like, you know, as a regulatory consultant, or as a… you know, quality person, or as whatever, a software development firm, it would be really great if I could do this here. Because our goal is to bring more and more… Consumers of this content on here, like, to provide them value, and then to provide the experts value as well.

Right. The way I look at this is that… You know. it's just a really nice place, especially, like, there's a place like all the videos and all the slides, and if I travel and I miss one of the Fridays, I can easily go back and still try to go to YouTube and find it, or wait for the email from Sean. I think this is just, very valuable to me, and I was shocked to hear only 38 paying members for MLVX. You know, I mean, when Sean gave out the 20% code, I signed up right away. The way I look at it, you know, there's no free lunch. We all try to make a living, and that's why I don't mind paying, as long as I learn something, and I did learn

a lot. for the last, what, two, three years? Because I joined this at the very tail end of Joe's, MN era, so I don't know how to better say that, but, you know, I just really enjoy this learning experience, and also with the 100 experts, and I can just click anybody if I need something, and for example, because of this learning session. I got to know Michelle a lot. And, now I'm her client in a very small capacity, but when I need something, I reach out to Tia, I reach out to Michelle, I can answer right away. So, you know, I think this is a wonderful thing as a community. So, you know, I would encourage people, if you could.

you know, just be a paid member to keep this community going, because I would hate to see the thing eventually fall apart if there's no financial backing of it. You know. So… I hope nobody hates me. I don't know.

I could kiss you right now. No, I appreciate you saying that, because I think that's all right on. And I think, you know, honestly, like, this isn't gonna… this isn't meant to be, like, some, hey, Sean and I are gonna get rich off of this, because it's not… we're not charging a lot, and there's not that many medical device experts out there, but you're right, like, there is a cost to this, and at some point, like, you know, we are putting a lot of effort into it. So I appreciate you saying that. And also, just, there's another feature that we haven't tested too much yet, but the idea here is gonna be that you can also ask questions, and I don't even know if I've tried this yet.

But, like… I don't know, let's think of a good question. Mom?

18, and 6…

But usability. Usability.

Is there another question? Okay, what? Okay, got it, got it, got it. Let's try this one. My startup is building a SAMD product. And we have a question on usability. This is very risky, because I don't think we've… this would be the first question. Do we need to do a… formative, usability, study if our… is Class A software. I don't know. I'm big on punctuation, and I guess I can tag it… tag it, like, 510K, and I'm gonna tag it… Usability, and… is there software on here? Softer, okay, cool. I'll pose the question.

See if this works. Okay, cool. So it worked. And so, presumably, as experts, you'll be able to come on here and… Post your answers. So… I'm actually not that sure. I think technically no, right? Carrie, you don't absolutely… formative usability studies aren't required usually anyway, right? Just to sum it, I'm not a usability expert.

Great.

Go ahead, Michelle. Go ahead, Danielle.

No, I was gonna ask, I'm intrigued by what you just said. So, if the answer is not there, an expert is going to provide the answer. The only question that I have about that is… How do we know, or who does any sort of certification or validation, like a second-person signature type of approach, that basically says, that statement Made by the expert is actually the… let's say the most correct. We don't have to say it's perfect, but the most correct, or a reflection of the current expectation of a regulator. How do we do that? Because.

Yeah.

as much as, you know, I want to put a label of expert to a lot of my friends.

Yeah.

I don't necessarily… trust their analysis in some cases, so… Nope. So, I wanted to hear your perspective on that, because then… Now, this is written as the expert opinion, and it may be off the mark.

That's right. So, I mean, the… the… it's… I'm kind of a free market, you know. both an economics and ideas type of person, and, you know, what I like about this is that, hey, look, it all worked. So now there's a question, and, and then there's an answer. And, like, I don't know, my answer was pretty… vanilla and not that great, but anybody can answer, right? So now Carrie can jump in here and go, like, Jose doesn't know what he's talking about, you know, of course you need formative usability study, or, like, whatever. And I'm already, as I'm doing this, I'm already thinking of… features that I want to add to this. Like, I'm thinking what would be great is, like, you know, you should really ask, and then to put in an add sign, and be able to have a list of people, and type, like, you know, Carrie, and have her profile come up and link her, and then she'll get a message. Like, that's not built in yet, but I'm already seeing, like, hey, that would be a cool feature to have, because

I might see a question, I might be like, hey, this is perfect for Daniel to answer, you know, so I might give a little answer, then tag you, and then you'll get a message saying, hey, there's a question here that I think you'd be the perfect person to answer. So, you know, we've done a little bit of this in Slack. Where, in the… there is a Slack, I don't even know, I haven't been in there for a little while, but there used to be channels for different categories, and so you might get a message, but this is more accessible to other people, like, anybody can really come in here and ask questions.

And, yeah. So, to answer your question, and we can also do things kind of like Reddit, where right now we do have the ability to flag a question, but you can also maybe have up and down votes, right? And so, whoever gets the most upvotes would show up farther at the top. But you can always come in here and go, like, hey, Jose's answer's not great, you know, love the guy, but this isn't his area of expertise. Don't trust him on it. And at the end of the day, like, you know, it's like with everything else in life, you get what you pay for, right? So, you're asking a question, you're getting free responses, you're not going to have a guarantee, but what you can do is get a question, and I'm seeing another thing that's not on here and I want to add, which is you should be able to click on my link and look at my profile.

And then click on my LinkedIn and go, like, should I trust Jose's answer on usability? Maybe not. So, trying to give those human touches to trust, which I think is what it boils down to. Yeah.

I think what's also interesting about the tagging thing is that that sort of leads to, like, lead gen, in a way, and You know, one of the things that I've struggled with as a marketer is, like, yeah, it's great, we do these webinars, and 900 people register, and 500 people show up, right? And then how many leads does the client actually get, and how many people are actually willing to have a call with them? But if you look at… if you look at, Well, here's a good example on the screen, is, you know, Claramed. and then the ability to click through on one of their profiles, and it leads to a download our usability kit, you know, as the call to action. And so the person downloads the usability kit, and they see at the end of the usability kit, hey, book a free one-hour call with one of our experts, right? So now you've created a channel for this to happen.

So, as you develop this content, you're just saying, okay, I didn't just do a webinar, I created a workbook, or I created a checklist, or I created a how-to guide. and, a white paper, for example. And so all these things, then, oh, you want to download the slides, or you want to do this, you want to do that, you know, you create a channel for opening up a discussion. And I think the thing that… AI is never really going to be able to do is, replace human conversations. So, the moment that I actually pick up the phone, and I call Carrie, or I call Elon, you know, that's where the magic happens.

And that's the point that we're trying to get to. The point is not to create another platform. The point is not to… Do all of this… long run for no outcome. It's… it… the point is… I know what everyone on this call wants, is more customers and more money. Okay? And so, my whole existence has been around engineering, getting people to talk to each other that don't want to talk to each other. Okay? And taking complex content that, on the surface, is boring and hard to listen to, and making it interesting. Okay? That's marketing. Marketing is taking information and transforming it

into knowledge. Right? Into something useful. And so the point of all this is taking your knowledge… it just gives you a tool for taking your knowledge and transforming it into something very, very useful. And from that, hopefully. You know, it will lead to more business. Because when I tell you that there is a tsunami coming for knowledge workers at this point, and that the way that people consume information and the way that people work with consultants. You know…

I'd brace myself if I were you, because the things that I'm reading are pretty terrifying, you know? And I am trying to build a platform that's looking 2-3 years into the future, not looking at, well, in 2015, we used to sell courses on our website. That's great that you used to be able to sell courses on your website. That's no longer the case. You know? And the people would…

By the way, like, on the flip side of this, too, like, there's obviously on the expert side of it, we all want new clients, new business, and all of that. But, you know, having been a startup founder myself, and having lots of clients who are startup founders, I got a soft spot for them. You know, when you're starting a company, and you're trying to build out your, kind of, your team of experts around you, you know you can't hire people all over the place. having a place where you can go and just see a list of them, and just going, like, okay, I need somebody on the cybersecurity side, whom can I reach out to, where I heard that I should be, you know, I want to think about, like, the EU, and how do I get out there, and then I can come in here, and in one place, see them, see their content, go, like, I don't know, T-Bor puts out a lot of content, we haven't uploaded it yet.

But once we do, like, somebody comes in here and they see he's consistently putting out really good content on a topic. you know, if I'm a founder of a company, I'm going like, hey, this is somebody I want to reach out to. So it's, As… as, like, on the other side of this, not just the experts, but on the… startups and then medical device manufacturer side, I think it's a great resource, you know, to just have a single place where you can not only see different experts in different fields, but see how they think, see what kind of quality content they put out, and it does place the onus on us, on the other side as the experts, to create good content, right?

I mean, still, it is true, I think, that content is king. You know, you're ultimately trying to build trust. And I always say trust comes in two flavors. There's a trust of, like, can I trust this person is ethical? And then there's, can I trust this person is competent? And, you know, here, you're able to show that competence through kind of like a show-don't tell, right? Like, it's not just, hey, look at my, you know, look at the three-letter acronyms after my name. It's more like, hey, look at my thought process on pre-submissions, or my thought process on threat modeling, or whatever. So… I know we're over time, Sean. I don't know if you had any last things.

There were a couple people that had questions that didn't get answered to try to follow up with you. Individually. I want to thank everybody for your time and your attention today. Stay tuned for more details. Again, as Jose said, you know, if you share this with your network, you know, and we'll give you a way to share it with your network, you know, you will get… what is it?

3 months free of any subscription that you want.

Yeah.

I'm the expert.

So you can't beat that with a stick. And, thank you all, Thanks, Jose. Working with Jose is amazing.

Oh, thank you.

Excellent job, guys. I've got a quick question. This is the first invite I've actually gotten into my email. Obviously, I've been kind of busy, and I haven't been able to attend Fridays. Are we gonna start Fridays up, and then were we gonna get invites directly to our email? Because I can't always find it on LinkedIn, so that's kind of one. are the reasons I have not attended as much. One, I've been busy, but two. It's hard to find the invite for me. Are you guys gonna start to, like, invite it to people's emails again? Because this is the first one I've gotten, and that's directly to my email.

I… yeah, we've had some problems with Zoom, but I'm just gonna have a list of people that get added as panelists regardless, because the invites aren't getting through, but the panelist invites are.

Okay.

If you want to be a permanent panelist, just… Ping me, and I'll add you to the list.

I want to be a permanent panelist. I can't always intend… but yeah, that's one of the reasons I can't find the invite, and so if I want to attend, I gotta make it easy to find it, and unfortunately, it's been kind of challenging for me.

Sorry about that.

No, I mean, in all honesty, it's a 50-50. I was completely swamped with QMSR, so half of it was my fault, and the other half was I wasn't getting the invite, so it was not… so it wasn't totally you guys.

Welcome back to the land of the living, by the way.

Thank you, Al.

Nice to see your smiling face.

raised terribly.

Yeah, so we're gonna figure out a way to publish the, MLV100 this weekend, and make it shareable. I'm, you know… I don't want to publish a 25-page document that gives people no reason to come to the website, and I don't… and I don't want to just publish a list, so it's like, ugh, how… you know, any ideas, let me know. But I'll figure it out. Yeah.

Got everybody picked, though. Yeah, I mean, that's one of the things you see right there, is… like I said, most of the people on the network right now are the MLV100s, there's only 116 experts, right? So there's just a few who are not. But right now, this is kind of the yearbook, and if you've got this little tag, it means that you were selected. And in your profile, there's a little blurb as to, like, why you were selected for the MLV100. So, congratulations if you're on this list. And, sean will also share something else.

That's it from me.

That's it? Great.

cute.

See, everybody.

Thank you.

Thank you. Nice weekend, everyone.

Everyone.

Otherwise, bye-bye, thanks.