Pod Insights Practical AI Using edge models to find sensitive data About this episode We’ve all heard about breaches of privacy and leaks of private health information (PHI). For healthcare providers and those storing this data, knowing where all the sensitive data is stored is non-trivial. Ramin, from Tausight, joins us to discuss how they have deploy edge AI models to help company search through billions of records for PHI.
Sponsors:
Neo4j https://graphstuff.fm/episodes/2023-finale-llms-and-knowledge-graphs-throughout-the-year?&utm_campaign=UCGenAI&utm_content=AMS-SrDev-ToFuDev-UCGenAI-Audio-None-GenAI1-GenAI-NonABM&utm_medium=Audio&utm_source=PracticalAI&utm_justglobal= – Is your code getting dragged down by JOINs and long query times? The problem might be your database…Try simplifying the complex with graphs. Stop asking relational databases to do more than they were made for. Graphs work well for use cases with lots of data connections like supply chain, fraud detection, real-time analytics, and genAI. With Neo4j, you can code in your favorite programming language and against any driver. Plus, it’s easy to integrate into your tech stack. Backblaze https://www.backblaze.com/cloud-backup/personal/landing/podcast/practicalai – Unlimited cloud backup for Macs, PCs, and businesses for just $99/year. Easily protect business data through a centrally managed admin. Protect all the data on your machines automatically. Easy to deploy across multiple workstations with various deployment options. Featuring:
Ramin Mohammadi – GitHub https://github.com/raminmohammadi, LinkedIn https://www.linkedin.com/in/ramin-mohammadi-mlDaniel Whitenack – Website https://www.datadan.io/, GitHub https://github.com/dwhitena, X https://x.com/dwhitenaShow Notes:
Tausight https://www.tausight.com/Upcoming Events:
Register for upcoming webinars here https://practicalai.fm/webinars!
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:06 Speaker 00 Welcome to Practical AI. If you work in artificial intelligence, aspire to, or are curious how AI-related tech is changing the world, this is the show for you. Thank you to our partners at fly.io, the home of changelog.com. Fly transforms containers into micro VMs that run on their hardware in 30-plus regions on six continents, so you can launch your app near your users. Learn more at fly.io.
0:43 Daniel Whitenack? Welcome to another episode of Practical AI. This is Daniel Whitenack. I am founder and CEO at Prediction Guard, where we're safeguarding private AI models. And I'm really excited today because I'm joined by a friend who I've had the pleasure of getting to know a little bit over the past weeks in the startup community, joined by Ramin Mohammadi, who is the AI and ML Lead at TauSight and also an adjunct professor at Northeastern. Welcome, Ramin.
1:13 Speaker 02 Hi, and thanks for having me.
1:15 Daniel Whitenack? Yeah, yeah, it's great to have you here. It's been cool to visit the Boston startup community a couple of times and participate in a few events together. I've been really fascinated to hear about some of the things that you're doing at TauSight, so I'm excited to dig into those a little bit. I'm wondering if you could share a little bit with us about kind of your really thinking deeply about the intersection of AI and privacy, but specifically as related to privacy and personally identifying information, but also personal health information, so PHI. So TauSight is thinking deeply about how companies are handling very private,
1:58 Daniel Whitenack? sensitive data and knowing about where that data is, which is actually a huge problem. I'm wondering if you could talk a little bit about, I was kind of shocked when I heard one of your presentations and you were talking about just the size of this problem and the scope of this problem related to PHI. So could you give us just a little bit of a sense of what PHI is, why companies handling PHI is kind of a problem, and some of the challenges related to that?
2:29 Speaker 02 Sure, I can do that. So first to give introduction, what's a PHI or Personal Health Identifiables? So based on the HIPAA rule, there are 18 identifiables which can lead to identify an entity or a person within a healthcare organizations. And these informations are valuable and being targeted by hackers. One of the reasons is that they have high value because they contain your sensitive personal information such as medical history, social security number and insurance details, which makes it very valuable on black market.
3:11 Speaker 02 They also use this for monetary gains, so hacker can sell this stolen PHI to criminals who use it for identity theft, insurance fraud, or other legal activities. They also use it for exploitation and extortion. Basically, they use this stolen health information to use for blackmailing and blackmailing individual organizations. So 133 millions healthcare data was breached in 2023, which means one out of three Americans' life was affected. This means about 160% increase compared to 2022 and about 240% increase since 2018.
3:54 Daniel Whitenack? So like for our listeners, if at least if they're in the US, one out of every three of those listeners has had some portion of their health information exposed in some type of breach. Is that right?
4:08 Speaker 02 That is correct. That record came out in 2023.
4:13 Daniel Whitenack? Yeah, yeah, that's insane. And you mentioned hacking. How is this data being breached, mostly by hackers or by sort of mistakes of organ? What is the combination of ways that this data is getting exposed?
4:27 Speaker 02 That's a great question. So actually, based on the report on 23, 78% comes from hacking of the network storage where the data resides in healthcare. And there are a small amount, which is about like a two persons, happens by when someone stole a laptop, for example, and the laptop contained PHI or sending an email via email or basically phishing emails, stuff like that. So there's like a breakdown, but majority comes from hacking, seventy eight percent.
5:00 Daniel Whitenack? And are these companies sort of mostly healthcare companies or like who has this data and how is it given?
5:09 Speaker 02 So that's, that's really interesting because technically healthcare organizations like hospitals are only, I think accountable for thirty percent of these incidents and the remaining seventy percent happens by hacking through their business partners, like third-party organizations that they have some sort of softwares or they storage for keep tracking of the medical data. But at the end, the cost comes to the healthcare organizations.
5:38 Daniel Whitenack? So that cost, what, what are the, like, what happens when this data is breached? What is the, what is the bad case scenario, I guess?
5:47 Speaker 02 Sure. So let me first tell you that how overall costs of this has been. So healthcare service security has spent about twenty eight billion over five years spent. And we are still not able to predict the PHI. And the way that it is that when an organization getting hacked or there's some data breached, depending on the states, there are some cap. If, for example, if you have been breached and you have lost more than five hundred entities or life, basically data, you need to go public about it. And also you will basically get sued
6:26 Speaker 02 and also you need to also get fined. So we have this wall, which we call it the wall of shame, unfortunately. And government posts the names of the organization where they got hacked or they lost basically PHI data. And this is this wall being constantly updated. And the last thing that any CISO or CIA wants is that to see their name on that wall.
6:53 Daniel Whitenack? Yeah, and their brand is hurt by that. But also, there's fines, right, for this sort of breach.
7:17 Speaker 03 What's up, friends? Do you remember when ChatGPT launched? I do. It felt like the LLM was this magical tool out of the box. However, the more you use it, the more you realize that's just not the case. The technology is brilliant, don't get me wrong, but it's prone to issues like hallucination on its own. But there's hope. There is still hope. Feed the LLM reliable current data, ground it in the right data and context, then and only then can it make the right connections and give the right answers. The team at Neo4j has been exploring how to get results by pairing LLMs with knowledge graphs and vector search. Check out their podcast episode about LLMs and knowledge graphs throughout 2023
7:59 Speaker 03 at graphstuff.fm. They share tips on retrieval methods, prompt engineering, and so much more. Don't miss it. Find a link in our show notes. Yes, check it out. graphstuff.fm, episode 23.
8:28 Daniel Whitenack? Could you talk a little bit about, like, let's say that I'm a healthcare company? And I want to not be on that wall of shame. And I want to do the best practices and all of that. Like, what is the reality of? You know, I know that, of course, you have been thinking very deeply about solving these issues with AI and machine learning or some related issues. But let's just say that doesn't exist. Like, what are the choices that a company has in terms of? And what are the challenges that they face in terms of securing this data?
9:05 Speaker 02 So currently, healthcare organizations have a series of tools. Some of them, like maybe four or five tools that they're going to do the same task. And the way that this traditional tools work is that they have a series of patterns, like a reject, for example. And when someone tries to download the data, if this matches, for example, with the some rejects that they have, it says, hey, you transferred PHI, or hey, you downloaded PHI. So those types of files, which they're coming directly out of EMR, electronic medical records, are easy to detect. The problem is with what we call dark PHI,
9:48 Speaker 02 a PHI that resides on your network, on your machines, but you are not able to detect it because the patterns that you are using are basically not capable of detecting that. You know, we have, for example, organizations that have millions of patients in your network. And I don't know if you probably have written the reject before. Reject is always as good as the person who's writing it.
10:15 Daniel Whitenack? Yeah, yeah. And it also, what is that meme? It's you decide to solve a problem with reject and then you just end up having another problem, which is reject.
10:26 Speaker 02 That's correct. So current from the prospects that we have talked with is that updating these rules is costly and requires a dedicated engineers or IT professionals. And no one likes to write.
10:41 Daniel Whitenack? Very true.
10:42 Speaker 02 At least I can say that. So that's a problem. They have tools in place, but no, it's incapable of solving the problem.
10:53 Daniel Whitenack? And when you say dark PHI, is this, I'm assuming there is like, okay, you might have a red X for a social security number or something like that. But if I just think of like a doctor recording a dictation of a patient visit or something like that, there's a lot of natural text in there about diseases and all of those sorts of things. So is that more natural text sort of health information? Is that what leads to kind of the dark PHI? Or does it also have to do with, you know, oh, it's easy to detect in this file format because I know the the pattern I'm getting, but then someone scanned
11:34 Daniel Whitenack? in this document and it's in like a PDF or something and my script doesn't know how to scrape these different data types. What could you kind of go into? I'm super fascinated by this idea of this dark PHI sitting around.
11:51 Speaker 02 So the first thing is that, point out that is 80% of healthcare data are unstructured. Unstructured means that from image to audio transcript, like all sorts of PDFs. And what we have seen also on the healthcare, it's a variety of data extensions. So you will be surprised that the, you will see file extensions that they don't exist. But what the clinicians or researcher do is that when they edit a file, they put dot the last name.
12:26 Daniel Whitenack? Oh, geez.
12:27 Speaker 02 So you, we, like, I think in the last study we did, we found 8,000 extensions on our basically prospects environment.
12:38 Daniel Whitenack? So literally the personal information is in the file extension?
12:42 Speaker 02 That's correct.
12:43 Daniel Whitenack? That's great.
12:44 Speaker 02 Personal extensions, personal information could be in the extension, but also random file extension that they might use in order to
12:54 Speaker 02 Basically bypass some rules.
12:56 Daniel Whitenack? Oh, okay, I gotcha. So they're doing a workaround because there's like this annoying tool that prevents this data from being transferred around is blocking this file type. So if I just change the file type.
13:09 Speaker 02 That's correct. So one thing also here to point out is that healthcare lives on data. Clinicians need to access that data and you you should not stop them basically from doing the job. You just need to have a better way to detect and basically maybe, for example, encrypt the file. So if someone else has stole that file, they cannot open the file. Our goal here is not is to prevent clinicians from doing something, it's just to make it more secure.
13:41 Daniel Whitenack? Yeah, and maybe that starts to get a little bit to kind of transitioning to how AI and machine learning fit into this puzzle. Before we exactly describe how it does fit into that puzzle, I think there might be a lot of listeners out there that are very intrigued by what the challenges might be of applying AI or machine learning in the context of healthcare. Like, what are the unique challenges if if you're a data scientist and you're, you know, building a model or wanting to use an LLM or building your own model to use in a healthcare context? What's unique about that context that makes it more
14:24 Daniel Whitenack? challenging to maybe maybe it's on the deployment side or the model building side? What are some of the challenges related to working in in healthcare specifically with this technology?
14:35 Speaker 02 This is an interesting question as someone who's keen on ML apps.
14:40 Daniel Whitenack? Yes, exactly.
14:42 Speaker 02 Say the main challenge is the whole project. But for us, these challenges are a bit more than some other AI technology due to the space which we are in. For example, we don't have access to real patient data to train our model. And no healthcare organization will agree to let you use their data. That's it. So.
15:05 Daniel Whitenack? And probably if one of them agreed to it, I'm guessing you couldn't use that same model for a different organization because you trained on specific data that's sensitive for one organization. Is that right?
15:16 Speaker 02 That's awesome. Correct. We have a huge data heterogeneity problem. And that comes like, for example, one organization's, it's for cancer organization. The other is like a dental organization. These datas are different. Other challenges that I can see is that you also cannot collect or transfer any data to the cloud. That means everything needs to happen at the edge. Data labeling is highly difficult. Even human level performance has about eight to 10% labeling error for detecting PHI's. Again, for example, you have lots of types and extensions.
16:00 Speaker 02 Data normally contains bias. Certain demographics have higher amount of data than others. Model development is confined by first model performance and then optimization metrics. And you know, model deployment on the edge has its own difficulties, which we can talk about later. I think lastly, unsupervised model monitoring makes it more challenging to detect drifts.
16:31 Daniel Whitenack? And just to kind of define a couple of those things, when you say the edge, what is your, because people might have different definitions in their mind of whether that's a, you know, some staff member's laptop or a desktop in a lab or something like that, or like a phone or a microcontroller somewhere, like people have a range of that. So in the context of healthcare, what is the edge environment?
16:54 Speaker 02 Edge could be from laptop that Kalisha is working with, from the desktops. Could be the tablet, for example, that you're using and could be the server storage, basically.
17:07 Daniel Whitenack? Gotcha. But all sort of onsite with some healthcare data center or where staff are working onsite?
17:15 Speaker 02 That is correct.
17:16 Daniel Whitenack? Gotcha. And when you say unstructured model monitoring, what do you mean around that? So what are you monitoring for? You mentioned drift, so I'm assuming that that has to do with like this PHI might also be changing itself in terms of what it's like there's a new form type or there's a new new thing that starts being collected. Is that is that what you mean by model monitoring and that drift element?
17:42 Speaker 02 One of the things that we will say is that the change in the distribution of the data as you scan across different groups within the same whole organization. For example, there's a group for radiology versus there's a group for like a normal PCP. And the datas have different distribution and we need to be able to detect any drift in the data distribution as early as possible. Sometimes you also might find something like concept drifts where it's more like the contextual, maybe a file that under certain scenarios considered as
18:22 Speaker 02 PHI. In some scenarios, it's not actually PHI. There are some rules over here, which makes it more difficult.
18:33 Daniel Whitenack? Yeah, yeah. Contextualization, I guess, is a challenge. Interesting.
18:46 Speaker 03 What's up, friends? I love BackBlaze. I'm happy to have them as a sponsor. BackBlaze makes backing up and accessing your data astonishingly easy. This is a service I personally use. Go to backblaze.com-practicalai. You get unlimited cloud backups for Macs, PCs, businesses for just $99 a year. You can easily protect business data through a centrally managed admin, protect all the data on your machines automatically, easily deploy across multiple workstations with various deployment options. You can add on enterprise control, including granular access permissions, advanced single sign-on, group management controls and compliance support. They even offer multiple restore options, including rapid recovery in the event of data
19:31 Speaker 03 loss or ransomware. That sucks. You can access your backed up data from anywhere in the world using their web app or their iOS or Android app. You can even restore by mail. They'll give you a hard drive with all your data shipped to your door. You buy a hard drive, restore, send the hard drive back within 30 days and get a full refund. Get one year file retention and version history. Over 55 billion with a B files restored for customers so far. Visit backblaze.com slash practically eyes. They know where you came from and continue to support the show. This is a service obviously recommended by me, but also by New York Times, Inc. Magazine, Macworld, PC World, LifeWire, Wired, Tom's Guide, Nine to Five Mac and just
20:16 Speaker 03 so many more. You receive a fully featured or no risk trial at backblaze.com slash practical AI. Again, their support in the show. Go there, play with it. Start protecting yourself from potential bad times. Start today.
20:43 Daniel Whitenack? Yeah. So maybe we could get a little bit now kind of into some of how you've been thinking about and approaching this problem and thinking about it from the the TAO site perspective. So in the context of this edge environment, in the context of this unstructured data, in the context of the constraints that we just talked about, how did you you and your team specifically think about applying AI and machine learning in the context of detecting PHI and maybe also like, what is your goal here? Is your goal to stop breaches? Is your goal to provide sort of insights about this PHI? How did you decide on what the main problem
21:27 Daniel Whitenack? is you wanted to solve and why AI or machine learning was relevant to solve that problem?
21:33 Speaker 02 I'd like to actually first tell a short story.
21:36 Daniel Whitenack? That'd be great. Yeah.
21:37 Speaker 02 Six months ago, I was on a going to tell the AI summit in Austin. I got to the airport and I was passing the TSA pre where the TSA agent asked me to check my bag. It turned out that the machine picked up on this pre-workout container which I had in my bag. The agent used this device on the box and the result was positive. He was like, oh man, I need to call for this special unit to come and check this. Oh, geez. I was like, sure. Then this special unit with something like a hazmat suit. Oh, geez. They came and used a kit with bunch of different reagent to more specifically test my pre-workout.
22:21 Speaker 02 They started sampling from the pre-workout and added to bunch of these test tubes. Long story short, the agent was like, yeah, you're good. He was like, this happened to my cousin also, this pre-workouts are causing false positive. So we are like that special unit with bunch of different model trains to find and protect the dark PHI. While the current tools in the market are like the first and the second machine, which leads to false positive or unknown false negative. At our side, we do see this problem as a personal problem. It's our PHI that's being targeted. And clearly, the current tools in the market, they cannot protect it.
23:06 Speaker 02 So HIPAA security rule says that you must do a complete and true assessment of all your risks and vulnerabilities to EPHI. It is so fundamental to what we need to do. And AI is such a critical piece of taking advantage of the newer technology around. Solving what used to be a labor-intensive problem that are could be much easier if you can define the scope of the problem and you can have machine learning models which can run effectively and accurately in a calibrated manner. That's what we do. We take advantage of the AI for to find sensitive data.
23:47 Speaker 02 And I think we get to the point that where risk and threats and vulnerabilities are going to be detected at the age using the AI. As opposed to gee, I have all these heuristic rules, which is how do we lots of stuff today when it comes to recognizing patterns. So at our side, we use AI, for example, to recognize when the sensitive data is in unstructured content. It doesn't require us to say, hey, there's a keyword here, there's another keyword here. That would be how you do heuristic programming. Combination of these three words must mean this, combination of this four must mean this. Those, you never get to do all the rules, you will never get to the variability you need.
24:31 Speaker 02 So in our model, for example, one of our model with about fifty million parameters model that can be set to recognize this stuff. You will never in years of programming get that much logic into your rigex, right? So now, the other main factor is for us is to ability to run this models right at the edge, where the data is being created, emailed, printed, copied or faxed. We bring the AI to the data rather than taking the data to the AI, which most of the current AI solution do that. By doing so, we can ensure that our data is always protected, agnostic of hardware spec or network connections.
25:16 Daniel Whitenack? Yeah, so all of what you said makes a lot of sense in terms of the approach and how you're applying AI and machine learning, but also in my practical sort of data scientist mind, I'm like, oh man, that's really, really difficult to sort of have these deployments of models, especially against sort of heterogeneous types of data, run them on the edge, run them across a diverse set of hardware. From your perspective as a practitioner, where do you think was the, the most challenging of those issues? Was it having to do with like the deployment targets and the diversity of those?
25:56 Daniel Whitenack? Was it having to do with the, the types of models that you could or couldn't run in those edge environments? That it have to do with the actual training and labeling of the data? I imagine all of those are really difficult problems to solve and you had to tackle all of them. But what were you maybe, what was some of the, the hardest problems to solve with respect to those things?
26:20 Speaker 02 I definitely will say the first, but the most challenging problem is the data labeling and data creation, because we don't have access to real patient data. So we need to create our own curated data set. Which we need to ensure that we don't introduce bias, creation bias in that data. The other thing comes around the model training. So our solution need to be able to live alongside other programs that are running on a given machine within a certain performance boundaries.
27:00 Speaker 02 One example I give you is that IT has set of rules. If there's an application surpasses certain memory or CPU, it will block that application. So you need to be sure that all these ML models that you have or this ML pipeline that you have always remains below this basically boundary.
27:22 Daniel Whitenack? And is that because like these are essentially, I mean, I might say mission critical, but these are sort of life critical systems, right? Like they're using these to treat patients, right? So if they
27:32 Speaker 02 That is correct.
27:33 Daniel Whitenack? If you pull the memory and the thing stops working, then it's potentially a life threatening type of situation, or at least a very concerning situation in the healthcare context, right?
27:46 Speaker 02 That is absolutely correct.
27:47 Daniel Whitenack? Gotcha. Yeah. And in light of those constraints, of course, some people now might just say, oh, well, we've got all these L L Ms now and they're, they're great at doing all of these things. But I'm guessing a lot of those aren't sort of fitting for this sort of environment, these memory constraints. So where do you go with that? Where, what is it looking back to sort of traditional N L P sorts of things? Is it model optimization? Is it a combination of those? How are you balancing the constraints, but also kind of looking forward to the, these new generations of models and that sort of thing?
28:26 Speaker 02 Regarding the L L Ms, I was reading about this F I three by Microsoft, the small model.
28:33 Daniel Whitenack? Yeah.
28:34 Speaker 02 And even that model requires certain amount of core RAM or GPU.
28:42 Daniel Whitenack? Correct. Yeah.
28:42 Speaker 02 None of the healthcare organizations have computer with those specs. They all have like four gigabyte of RAM maximum and some legacy CPUs.
28:55 Speaker 02 The other problem with LLM is that it introduces some additional risk to the healthcare. Some clinicians or researchers, they're using tools like, for example, chat GPT to copy paste patient data to get some summary extraction, which it's not how it should be used. I know, for example, I know actually your company, Prediction Guard, you are trying to solve a problem like that.
29:25 Daniel Whitenack? Yeah, there's certainly a lot of people pasting things into chat interfaces. That's very concerning. I'll definitely say that. Yeah, for sure.
29:33 Speaker 02 That is great. Now, when it comes to model optimization, we take a series of approaches to be sure that our models are optimized for such an environment. This could be from knowledge distortion or student teacher networks, quantization and model pruning. We do take a combination of all of these to ensure that every model that we have lives within a certain boundary.
30:01 Daniel Whitenack? Gotcha. Yeah, yeah, that makes sense. So it's very important. And I'm guessing the model architectures and approaches that you, you can only go so far. It's not like you're going to take Llama 3, 70 billion and do these optimization techniques and fit it into four gigabytes of memory and run it on a CPU. So it's super interesting and I think, I don't know, what is your view as, as maybe you observe in the marketplace? People are exploring these open models, exploring bigger models. But at least in the space that you work in, the only way that you kind of move forward is with small models or customized models, optimized models. How do you view that kind of shifting into the future?
30:46 Daniel Whitenack? Do you think there will always be this sort of diverse set of environments in the healthcare space that you need to optimize models for? Will they eventually, you know, get over their hurdles of using kind of the cloud or large models? How do you see that developing, moving forward and into the future?
31:06 Speaker 02 A report by Schneider Electric indicates that currently 95% of AI workload operates on data centers.
31:15 Speaker 02 They have forecast that this number to go to 50% between edge and cloud by 2028. When you're monitoring the current developments in the market, you can see that most of the chip manufacturers are moving towards creating much stronger chips or machines where they allow to run the AI at the edge. For example, Intel's Meteor Lake or the AI PCC, right? But it will take quite a while for healthcare organization to have that change adapted. Because, you know, it requires budgets and I think healthcare organization, they go through
31:58 Speaker 02 machine update once every five years. I don't think they do it over all data machine, only maybe certain machines. But definitely I do see the future that you can bring much larger models right at the edge. But I don't think we're there yet, no.
32:17 Daniel Whitenack? Yeah, I appreciate that perspective because some people, I think, in our listener base, like they're constantly overwhelmed by this news about these new big models. But it's harder to get this sort of story of a practitioner on the ground, working with specific companies in certain constraints. There's still quite a diversity of constraints that practicing data scientist or AI engineer has to work within. So I think that viewpoint's very important. As you kind of look at what you've done with with TauSight and these tools that you've built in detecting PHI, helping companies know where their PHI is, reducing false positives,
33:01 Daniel Whitenack? figuring out how to run these models on edge devices and all of those things. Do you have anything that stands out in your mind in terms of, you don't have to mention specific customers or anything, but but success stories or really things that you're proud of, that you're glad that you've been able to be a part of in terms of helping protect this this PHI, any any sort of case studies or or use cases that pop into your mind?
33:26 Speaker 02 Yeah, I can give some example without naming anyone. But we have, we were in this meeting and this CISO was in the call and it's like, I had this laptops that was stolen and I don't know what's on that laptop. But because we have our software on that laptop, we kept an inventory. And it turns out the laptop contain lots of PHI's. So they never have that view on this type of scenarios that where the PHI is or who has access to it, but we can basically give you that.
34:05 Speaker 02 Oh, I was on another customer call and it was like, we are happy, you know, with the tools that we have. They have less false positive, but unknown false negative. And we are quite unhappy for writing the rules.
34:22 Daniel Whitenack? Yeah.
34:23 Speaker 02 You know, it takes us a while. But when you use our tool, there's no rules that you need to write. It's basically out of the box after you installing ours product. It will start scanning all the files and also monitoring what's happened on the machine. So if someone copying or pasting, and for example, PHI into an email or into another file, we can read that. We can detect that if someone faxing it, we can basically detect that. So this customer calls that they happen on, they also got to be positive.
35:01 Speaker 02 And sometimes scary for customers because, you know, we are able to find really detailed information around your network.
35:11 Daniel Whitenack? You're pulling the curtain back. There's work to do once you understand it. Yeah, well, maybe as we, as we kind of draw pretty close to a close here. As you are kind of plugged in both on the academic side, you're plugged in on the startup side with Tao site. You're, you know, investing in this healthcare industry.
35:34 Daniel Whitenack? From the perspective of AI and ML, what gets you excited as you look to the coming year? Maybe it's things with Tau side, maybe it's things more generally in the AI community. What are you excited about? And what do you think is some of the positive things that you're you're seeing kind of develop over the coming year?
35:51 Speaker 02 I think there are two things that I'm really interested. One is the development on these large models and the fact that they're getting smaller and smaller. I am looking for the day that I could work with those SLM or small large models and deploy those on right the edge. And I think the other thing that I'm quite interested and right now actively working on is the federated learning. I know federated learning is kind of in the background, not many company actively are doing it due to all the challenges that
36:30 Speaker 02 it has and also some security concerns. But for a domain like healthcare, where you cannot transfer data and you cannot see the data, I found that absolutely necessary that for your models to be able to train themselves and update themselves. So I think those are the two main things that I'm looking for the upcoming years.
36:53 Daniel Whitenack? That's awesome. Yeah, I'm definitely excited by both of those things as well. And I know Chris and I on this podcast have mentioned federated learning for for years. I hope that it kind of comes comes more to the forefront as people figure out paths to do this. I think that will be interesting. Well, Ramin, it's been great to have you on the show. I think we'll see each other again in Boston before too long, I think. But yeah, it was great to have you on the show and thanks for taking time out of out of your schedule to share some of these insights with us.
37:27 Speaker 02 Yeah, thanks for having me, Daniel, and see you soon in Boston.
37:32 Daniel Whitenack? Sounds good. See ya.
37:41 Speaker 00 All right, that is Practical AI for this week. Subscribe now if you haven't already. Head to PracticalAI.fm for all the ways and join our free Slack team where you can hang out with Daniel, Chris, and the entire changelog community. Sign up today at PracticalAI.fm/community. Thanks again to our partners at fly.io, to our beat freaking residents, Breakmaster Cylinder, and to you for listening. We appreciate you spending time with us. That's all for now. We'll talk to you again next time.
Previous Rise of the AI PC & local LLMs Next The perplexities of information retrieval Page updated October 26, 2026 · transcript and summary are machine-generated