The Security and Trust Outlook for 2026 (with Debbie Reynolds)

In this episode, Leonard is joined by our good friend and privacy and security maven, The Data Diva, Debbie Reynolds of Debbie Reynolds Consulting to get a feel for the outlook of security, privacy, and trust in 2026.

The security and trust landscape is constantly shifting, and the threats and risks are growing and accelerating as GenAI expands the fine line between digital marketing (surveillance economy) and the AI-augmented criminal economy.

Leonard and Debbie unpack the following:

💥 Where are we going with security & trust in 2026?
💥 Taller walls & bigger locks are not good enough!
💥 Is breaking down silos of data for AI expanding security risk?
💥 The neXt Curve AI Security Sandwich
💥 The missing focus on curation and knowledge
💥 Over-trust and over-reliance on GenAI danger
💥 The death of Metaverse & the rise of AI glasses
💥 The rise of the robotic privacy problem
💥 The outlook for trust in the era of deepfakes
💥 The AI chatbot for spam & agentic cyberattacks
💥 AI will only get better,... for threat actors too!

Hit Leonard and Debbie up on LinkedIn and take part in their industry and tech insights.

Check out Debbie Reynolds at www.debbiereynoldsconsulting.com. Also make sure to check out "The Data Diva Talks Privacy" podcast, celebrating its 1 millionth download! Congrats, Debbie!

Please subscribe to our podcast which will be featured on the neXt Curve YouTube Channel. Check out the audio version on BuzzSprout or find us on your favorite Podcast platform.

Also, subscribe to the neXt Curve research portal at www.next-curve.com and our Substack (https://substack.com/@nextcurve) for the tech and industry insights that matter.

NOTE: The transcript is AI-generated and will contain errors.

DISCLAIMER: This podcast is for informational purposes only.

Transcript

0:06

Next curve.

Leonard Lee: 0:09

Everyone. Welcome back to Next Curve's Rethink podcast series on security and trust, where we talk about the hot topics in the world of technology, privacy, security, and trust. And of course, all of this has to matter, and so that's why I have my good friend, the data diva. Debbie Reynolds of Debbie Reynolds Consulting and all her Fantabulous Glory. So wonderful to have you as always. And it's been far too long. we haven't really kept up with our cadence. I know that we're, we've both been busy. But, yeah, it's about time that we caught up because, we're kicking off 2026 and the next curve audience needs to know what the data diva is thinking about. What's the, what's top of mind. And so what we're gonna do, everyone is we're going to, talk about security and trust outlook. What is the outlook for 2026? But more importantly, what are some of those issues and risks? That, you folks out there need to be, aware of and, mindful of as you, saunter into this new era of Ag Agent AI and digital X, y, Z, which is getting more confusing and more convoluted every single day with. All kinds of problems that actually, Debbie, you and I have talked about for years now. Mm-hmm. but why don't you say hi to the audience. It's been a while.

Debbie Reynolds: 1:42

Hello, audience. Happy to see you. Happy to have my, very good friend Leonard, chatting with you. Always. I was, actually going but down memory lane, noting that you were, my third podcast guest. A Pine number three. Yeah. So we're now at like two 70 ish, something like that. We had over a million downloads.

Leonard Lee: 2:03

Yes.

Debbie Reynolds: 2:04

It's been cool.

Leonard Lee: 2:05

Congratulations. That is amazing.

Debbie Reynolds: 2:09

Yeah. I'd have to get you on my show so we can find out what the hell's been going on. For last five years.

Leonard Lee: 2:16

Well, that's, uh, maybe yeah, it's like a five year retrospective.

Debbie Reynolds: 2:21

Totally.

Leonard Lee: 2:21

First half of the 2000 twenties.

Debbie Reynolds: 2:26

Exactly.

Leonard Lee: 2:26

But, yeah, I'm looking forward to this. This is about time we caught up and hopefully we'll be able to do this regularly, but

Debbie Reynolds: 2:32

yeah.

Leonard Lee: 2:33

You know we're both busy. but we're always in tune with each other, you know? Of course, those of you who watch iot Coffee Talk. Debbie is an iot coffee talker, so, every week we're on, we do that, snow, rain, shine. That's right. Whatever. And, uh, Debbie's one of the crew and she always drops her wisdom in every episode that she, appears on. And so, um, yeah,

Debbie Reynolds: 2:57

Sometimes we go blue. Yeah. So yeah.

Leonard Lee: 2:59

And, we go crazy. That's like unfiltered.

Debbie Reynolds: 3:03

It is.

Leonard Lee: 3:04

We have to behave, right? We have to behave. But before we get started, remember to like share and comment. On this episode and subscribe to the Rethink podcast here on YouTube and on Buzz Sprouts. And take us on the road and on your jog and listen to us on your favorite podcast platform. And this, podcast is for informational purposes only. statements by my guests are their own and don't reflect those of next curve. We just want to provide this open forum for debate and discussion for all things tech and. Industry as well as, security and trust. again, want to congratulate you on a million downloads on your podcast. That's, quite amazing and it's been a privilege to watch you. grow your practice, grow your business, and, grow your imminence. it's wonderful to see that you are such the leader, the thought leader, and pioneer. And, you and I have known each other for now years.

Debbie Reynolds: 4:04

A long

Leonard Lee: 4:05

Amazing. Yeah. You know, what is that 290, 90 episodes back or something like that? Yeah,

Debbie Reynolds: 4:17

totally.

Leonard Lee: 4:18

Podcast at least, right?

Debbie Reynolds: 4:20

We knew

Leonard Lee: 4:20

each other.

Debbie Reynolds: 4:21

Yeah, we knew each other before that.

Leonard Lee: 4:23

Yeah, yeah, yeah. But you know what? 2026 coming outta 2025, just looks like a nutty year.

Debbie Reynolds: 4:32

Yeah.

Leonard Lee: 4:33

Undoubtedly you see a lot of the issues forming. Technology landscape, the geopolitical landscape, the regulatory landscape. it's just shifting Remarkably fast coming into 2026, what are those things that are top of mind for you? Let's systematically go down. Some of these, I think it really has to be across because you cover so many things from IOT to spatial computing, to, the regulatory stuff. You're looking at things from a geographical and regional perspective. So, what do we all need to, work on this year?

Debbie Reynolds: 5:07

Yeah, well, there are several things that people need to think about. I would say the thing that is concerning me. and I guess I'll give you an example. So, many moons ago, you remember when Target had that big breach, where some third party company or contractor had access to their system, and then they went into their system, they did all this bad stuff or whatever. Yeah. And so I feel like a lot of times the conversation around data. Trust is around, you know, we just need better locks and better doors. And I'm like, well, if they get into the door, what can they do? Right? So if you're. Really managing your data the right way, then it should be limits to what havoc someone can wreak when they breach your organization or when data leaks. And so I feel like we're not talking enough about that. We're just saying, let's just put more data in. Let's try to build bigger, taller walls, bigger locks. And it's like, that's not the issue in my view. So I think the data transfer is more about context, in my view, you know, what someone needs to see or what they don't need to see. And at some point I think we need to get to a point where we're thinking about. You know, we're architecting things from a software, hardware perspective to make it like really hard for someone to do something bad with data. And part of that to me is like less data.

Leonard Lee: 6:41

Yeah.

Debbie Reynolds: 6:41

Like being more Yeah. Specific around what you're doing instead of

Leonard Lee: 6:45

Yeah.

Debbie Reynolds: 6:45

Trying to do everything.

Leonard Lee: 6:47

Yeah. Well, you know, the interesting counterpoint to that is this whole movement due to AI training and just everyone trying to crack the enterprise AI code, looking at bringing down the silos of data, right? And silos, information, knowledge, Enterprise discovery. That's really difficult to do because this is not like R-D-B-M-S, you know, our role-based, access control is really hard to institute up and down the stack for a, an ai, application or a system. It's very difficult, if not impossible in many instances, depending on what you're trying to do. And this is something that you and I have talked about actually, for three years. because we actually were looking at the technology and we understood what are the limitations of vector databases, LLMs, you know, it's not just about hallucination. Everyone fixates on hallucinations, but there are all these security issues. But then at the same time, there is this counterpoint where I. You know, especially vendors are trying to push organizations to open the aperture for access to data, and that becomes really problematic. Right. And I think a lot of organizations are still going through that discovery process. So I don't know what your reaction to that is,

Debbie Reynolds: 8:19

yeah,

Leonard Lee: 8:20

it's interesting that you say that because yeah, we, we've been trying to lock down applications and there is this proclivity now. to share less, whether it's because of, copyright infringement or confidentiality, concerns, privacy concerns, but,

Debbie Reynolds: 8:37

mm-hmm.

Leonard Lee: 8:38

Yeah. You know, I,

Debbie Reynolds: 8:39

I think enterprises need to relearn or reeducate themselves about the way they're supposed to handle data. So the way it is now is like, let's create a new big bucket of stuff and just throw it in there and then hope we press some buttons and something magical comes out. And that's not the way it goes. It's like you need to be more circumspect about what you collect, why you collect it, why you have it. Because maybe this is a holdover from the big data days, right? Where it's like more data is better. It's like more data is like more risk, right? So it's like, why are you taking more risk? A lot of times the risks people are taking is like. When they look back, they're like, okay, we didn't even need that data. That data was not even useful, useful for us. So now we're in trouble some way. And it's like if you put a tighter rein on what you have and why you have it, and really ask those questions as opposed to just adding more data and just hoping that you can find some magical tool that's gonna help you.

Leonard Lee: 9:38

yeah.

Debbie Reynolds: 9:39

the bonkers days when people were trying to do enterprise search where they put an app on everything. They say, yeah, let's expose everything and try to search. And it was a disaster. Because people didn't understand what was. accessible.

Leonard Lee: 9:55

Yeah.

Debbie Reynolds: 9:56

within the enterprise, you know, a lot of this security biosecurity goes out the window when you do that. And so we're having like another moment of that where people throwing agents on stuff and it's like, but your data isn't, locked down. It's not secure. You don't know what's there.

Leonard Lee: 10:12

Yeah.

Debbie Reynolds: 10:12

So now you're basically, you're like stirring up all these. You know, all this data and creating all this new internal and external risk because you don't really know what you have. So I don't know. Yeah, it's kinda like, eat your vegetables, I guess that's what I'm trying to say.

Leonard Lee: 10:27

Yeah. And that was like, so I have this concept called the AI security sandwich.

Debbie Reynolds: 10:34

Yeah.

Leonard Lee: 10:35

I come up with the most ridiculous. Things, I have to admit that. but yeah, one, the lower half of the sandwich, the bottom piece of bread is, uh, data, uh, security, posture, management. And the first step of that is to really ca, you know, like take an inventory and catalog and understand, what, what data and information all the way up to knowledge that you have. And see, that's one of the problems I have with this whole AI discussion where it's fixated on data. When we think about data, we think about a certain level of information, right? And that's usually the lowest level. And when you talk cross domain about data, it means different things. But fundamentally, it's really uns synthesized information. It's just raw data that you're talking about. There's this whole hierarchy of, other levels. Ultimately, culminates in knowledge, a knowledge base. Right? And we should be talking about knowledge bases more, but a very fundamental first step is that understanding the posture of your data, information and knowledge and knowledge is a different beast.

Debbie Reynolds: 11:48

Yeah.

Leonard Lee: 11:48

and a lot of organizations haven't done that. that really is one of the first critical steps. And then there's that top part of the sandwich, which is your data loss protection type of stuff, where you're securing endpoints, you're securing access from an endpoint perspective. And then, you know, there's a lot of holes at that level, but That's all we have right now. Everything in between is like, it literally is like A-A-B-L-T made to order and with too many toppings. Do you know what I'm saying? It's like a

Debbie Reynolds: 12:20

mess. Yeah. Yeah.

Leonard Lee: 12:21

So,

Debbie Reynolds: 12:22

Maybe I'm bringing it back to the Matrix now. I feel like we're missing the librarian,

Leonard Lee: 12:27

the library. Exactly. Exactly. And then, you know, not all, I mean, not all data is good data. A lot of it's

Debbie Reynolds: 12:33

no

Leonard Lee: 12:33

Right,

Debbie Reynolds: 12:34

right.

Leonard Lee: 12:34

Yes.

Debbie Reynolds: 12:34

So we're missing that curation step. That's what we're missing. So think of going into a library and none of the books were organized or whatever, and people are like, oh. it doesn't work that way. So. it needs to, I don't know if the word is like gatekeeper, but someone there needs to be some level of curation. Yeah. That is not happening. So that's why we're seeing a lot of companies fail at these enterprise projects because they skip that part. They're hoping, oh, well the AI will curate. It's like, no,

Leonard Lee: 13:02

no,

Debbie Reynolds: 13:02

it's not gonna do that. It's just like a broom does not curate dirt on the floor. You know what? The data's not gonna curate itself. So there has to be some human

Leonard Lee: 13:14

right

Debbie Reynolds: 13:15

synthesis that occur somewhere. And a lot of times we're thinking, we'll just let AI do it. That just never, that just has never worked.

Leonard Lee: 13:23

Yeah. And you know what, I think one of the problems is like, you know, so we used to have knowledge managers in organizations when knowledge management. The craze, right? And document management, getting your stuff tagged and organized and discoverable and, you know, fully lifecycle managed. but then, they suffered what librarians suffer in your public library, right? they get disregarded, right? they're like treated like, some nerd that tells you to stay quiet and behave while you're browsing the rows of books. And, I don't think they got the attention. Or the respect within organizations for the important role that they would play after there was all this effort to actually get knowledge, organized and baselined, right? Because there's a whole life cycle that a lot of organizations don't do a really good job of, managing.

Debbie Reynolds: 14:19

Yeah, totally.

Leonard Lee: 14:20

And so, yeah.

Debbie Reynolds: 14:21

Well, right, so there is a problem within the organization, but I think also with AI now, it extends outside of the organization.

Leonard Lee: 14:29

Yeah.

Debbie Reynolds: 14:29

cause then there's a lineage problem with the data.

Leonard Lee: 14:32

yeah,

Debbie Reynolds: 14:33

the traditional idea of governance, it's like I put my blinders on about data until it comes into my possession. where it came from is. Very important.

Leonard Lee: 14:42

Yeah.

Debbie Reynolds: 14:42

That's an important part of the puzzle that a lot of companies aren't really thinking about, and they're just like, again, I think, and it's hilarious. I'm not gonna name the company, but like a big insurance company, a big one. They were like, yeah, let's just throw everything into like generative ai. I'm like, what? I can't do that. It's not like a database, there has to be, and that's probably another thing we could talk about. I'm seeing as you're seeing a lot of these generative AI tools being pushed into healthcare. Yeah. And it's very scary. So for me, I'm concerned about,

Leonard Lee: 15:17

yeah.

Debbie Reynolds: 15:18

Accuracy.

Leonard Lee: 15:20

liability.

Debbie Reynolds: 15:21

Yeah. Act right. you don't, oh, it sucks to be you. Oh, oops, we gave you a misdiagnosis'cause we didn't use the most accurate way of doing something. It's certain tools don't do. certain things, so yeah, like a hammer is good for a nail, but maybe a screwdriver doesn't help you.

Leonard Lee: 15:40

Yeah.

Debbie Reynolds: 15:40

Put a nail on the wall. So I feel like we're trying to put the wrong tools to the wrong things.

Leonard Lee: 15:46

There's a lot of. Corner cases being used to characterize large language models in these chat bots as being trustworthy and accurate. And I think that's seriously dangerous. quite frankly, that has to stop.

Debbie Reynolds: 16:04

Yeah,

Leonard Lee: 16:05

it really does. because I would hate to be a physician having to argue with a patient. and try to regain the trust of a patient who over relies on and is overdependent on, right? Let's say, a chat bot.

Debbie Reynolds: 16:20

Totally.

Leonard Lee: 16:21

That's a serious problem. that's overhead for the physician that actually makes their work. More laborious and when trust gets transferred from the physician who has a duty.

Debbie Reynolds: 16:34

Yeah.

Leonard Lee: 16:34

To be careful and, to, look after the health of their patients. Right? They have that obligation having to fight for the trust of their patient, with an AI agent of some kind, you know? and I don't think, I don't see that people have connected those dots and it's such an easy stream of dots to connect.

Debbie Reynolds: 17:00

Yeah.

Leonard Lee: 17:01

You know what I'm saying? It's remarkable.

Debbie Reynolds: 17:04

It is, it's very scary. So I think of course for people who build these tools, they love it.'cause it's like, okay, we sell this product to this healthcare company and it's like,

Leonard Lee: 17:16

yeah,

Debbie Reynolds: 17:16

this is all about, it sucks to be you type of thing. So if you make a mistake or use it wrong or whatever, the company, they could, they just say, is at your own risk. And I've never seen products being rolled out. In that way. Yeah. And I think that people almost assume that there is a level of safety there, so they don't understand that is, there is really a combinant upon them to decide what's the best way to use these tools.

Leonard Lee: 17:43

Mm-hmm.

Debbie Reynolds: 17:43

In a way that's helpful and not harmful.

Leonard Lee: 17:46

Yeah, Let's talk about spatial computing and XR for a moment, because I know that you've done some work there recently. and you're a privacy advisor in that regard.

Debbie Reynolds: 18:01

Yeah.

Leonard Lee: 18:01

so not the metaverse the hot things happening there because smart glasses, it feels like CES was like ages ago. Yeah. It just happened two weeks ago. It's hard to believe. but yeah, these, smart glasses are all the rage, at the moment.

Debbie Reynolds: 18:15

Yep. Yep. Everybody wants glasses. But we talked about this

Leonard Lee: 18:18

Oh yeah.

Debbie Reynolds: 18:19

Couple years ago, right? We talked about, and we talked about, remember we talked about the vision pro? Yeah. And we talked about all the patents they had, and I said, Apple's gonna do other products, which they're doing. That's exactly what they did.

Leonard Lee: 18:32

they have so much headroom in terms of technologies that they can now squeeze down into something. creepy.

Debbie Reynolds: 18:39

Yeah. I mean it's all about creating a deeper relationship, right? So yeah. Personalization, whenever they say that you should be, but who, you should be afraid when they say that, but, yeah, I think. From my perspective, and I think we've said this for years, my thing was when people are like, rah rah about the metaverse? I was like, to me, spaceman computing is not about you trying to go to a place quote.

Leonard Lee: 19:07

Yeah.

Debbie Reynolds: 19:08

To me it was about. How do you interact with your environment? Yeah. So it's not like, like an amusement park that you go to. It's how does it help me in my life? And so I think the uses that we've seen, we've seen some uses in education. We've, I've definitely seen some uses in in health.

Leonard Lee: 19:27

Yeah.

Debbie Reynolds: 19:28

But I think what they wanted is more of a consumer.

Leonard Lee: 19:31

Consumer, yeah.

Debbie Reynolds: 19:31

Rah rah. You know, like gaming, the gaming industry and stuff like that. And so that part didn't happen. And part of that is the expense, right? And then, you know, people have stuff to do besides buy very expensive glasses.

Leonard Lee: 19:48

meta made'em pretty cheap and, accessible. But that formula still didn't fly. And just so that we commemorate the, what looks like the death of the metaverse, I mean, basically scrapping that whole group, right. So, yeah. it looks like, Meta is pivoting hard toward smart glasses, which makes sense because not, you know, you're never gonna, one of the big misconceptions about vr, even Mr. to a certain extent, not as much, was that you're gonna have people walking around with these ski goggle looking things, in public, which you had some, you know, uh, I don't wanna call'em morons. They are, um, going around with, an Apple Vision Pro and driving their Tesla. That's not what they're designed for, right? that's not the use case, but that's exactly why for the ambition that I think a lot of these, companies had, that didn't fit the bill. It was Not going to happen. Right? Yeah. The gaming thing, you're fighting physics. Totally. You're just going to get sick. There's no way around it. It's like trying to solve problems that can't be solved.

Debbie Reynolds: 21:01

Right.

Leonard Lee: 21:01

Saying like, hallucinations with LLMs.

Debbie Reynolds: 21:04

Right.

Leonard Lee: 21:05

You can't solve it. It's just you are not gonna solve them. It's,

Debbie Reynolds: 21:09

yeah.

Leonard Lee: 21:10

It's just part of. The way of things, you know?

Debbie Reynolds: 21:16

Yeah, yeah. And then, you know, I'm seeing obviously people, companies are trying to double down on things that they think are more practical. So like, I don't know about you, You went to C Cs. Yeah. some of the reporting I saw coming out of there, I'm like, how many robot vacuums are you guys gonna put out? I was like, oh my gosh.

Leonard Lee: 21:35

Yeah.

Debbie Reynolds: 21:35

And to me that's like a practical thing that people could possibly use. But like it's, I want something, instead of telling me that my refrigerator's open is to close the damn door, that's what I want.

Leonard Lee: 21:48

No, but you know what, um, I think we, uh, we're, I don't know where I made this comment, but, um, my wife, my wife will kill me. If she ever watches this episode. But the convenience fosters, bad behavior. My wife is absolute expert on all these robotic vacuum cleaners. It's like every six months we have like the latest version, right. And my kids now have this, um, proclivity or a habit of just like throwing stuff on the floor because they just assume that the robotic vacuum cleaner at some point.

Debbie Reynolds: 22:30

Right,

Leonard Lee: 22:30

picking it up. So during most of the course of the day, because we have this thing running in the late evening, you have like food all on the floor. And I've actually had my wife tell me, don't worry about it. The robot will pick it up. It's just like this huge sandwich on the floor. I was like, honey, pick that up.

Debbie Reynolds: 22:53

Oh my gosh. Oh my goodness. That's so funny.

Leonard Lee: 22:58

Yeah.

Debbie Reynolds: 23:00

It's funny. It's funny. It truly

Leonard Lee: 23:03

is. Yeah, it is. I guess it's just a sign of how ridiculous things are getting. But yeah, I mean, there's a privacy aspects as well, because these things, they have lidars on'em now Totally span your entire, environment in real time. And it's like, okay, so what about the security, privacy and confidentiality there? let's say that now you start to use these robots and they're for rent. They're connected. You use it for your offices, cleaning your offices, maintaining your offices, and this thing's creeping around. You know, what we've seen with some of these companies? They don't actually have autonomous robots. There's some guy in a closet somewhere.

Debbie Reynolds: 23:43

Yep.

Leonard Lee: 23:44

Controlling it remotely. Right? So this is the, you know, big, big ups for 5G and wifi, but. It certainly isn't a statement for autonomous robots, Yeah, I don't know. Where are we with that stuff, because, you know, I haven't really looked. Very closely at the privacy. I mean, I can just assume and kind of project what all those issues are gonna be, but where are we with industry or regulators looking at this? and from a, consumer protection perspective, I guess that's really where you have to start, right?

Debbie Reynolds: 24:18

So we are not seeing any movement on a regulatory front, especially in robotics and stuff like that. I think mostly because a lot of the use cases that people think about mm-hmm. Like, you know, Neuralink or whatever, some of those, because they're related to medical, they're like, okay, they're. In the us they're hipaa, right? So there's a HIPAA protection there, but the issue is when these things become more commercial. Mm-hmm. but then all that in the US is more consent based. So it's like, again, it sucks to be you. Right? So, you know, this robot may fold your laundry and make you toast and stuff, but it also is tracking you. Everything that you do.

Leonard Lee: 24:56

Yeah.

Debbie Reynolds: 24:56

in your house or whatever. And so, you know, there's a big gray area there. Around that type of collection. But, there has been, a very interesting development that's happened recently around cars, but I think,

Leonard Lee: 25:14

yeah. No, but everyone, she is like the person. For privacy and security and vehicles. it's insane. She's done wonderful work there.

Debbie Reynolds: 25:25

Thank you, thank you.

Leonard Lee: 25:26

Yeah.

Debbie Reynolds: 25:26

well, you remember a couple years ago there was a big, uproar around GM and they're collecting car data, selling it to. Data brokers who were using it for insurance. And so, that was like two or three years ago that article came out in the New York Times. Well, now the FTC, federal Trade Commission, put out a consent order for gm. So they have to stop selling for five years. They can't sell any consumer data to, data brokers or consumer credit. Agencies like Lexus nex and for 20 years they have to have a robust privacy program, including showing people how they can opt in, opt out of things, not just checking boxes and stuff. So I think that's gonna have a huge, huge, impact not just on cars, but also like other IOT devices. So, yeah, they actually carved out, language around Car data or

Leonard Lee: 26:27

yeah,

Debbie Reynolds: 26:27

driver data or people data. And so I think that that blueprint could easily find its way into like other iot devices.'cause basically you're saying, you know, people who buy products that are connected to the internet, if you collect data about them, you have a responsibility to that person and you can't just take it and you have it for a secondary use. Without the person not knowing. And that's the whole data broker industry, literally.

Leonard Lee: 26:58

Wow. Yeah. And it seems, amazing that we're still talking about this'cause you and I have been talking about this for at least eight years now.

Debbie Reynolds: 27:09

Yeah. It's a long time.

Leonard Lee: 27:10

And we still have a long way to go with this whole opt-in.

Debbie Reynolds: 27:14

One of the big things that they did, which I'm happy that they did, is that, and this has been a, a argument that's been brewing for a super long time, which is a lot of device makers say, the data about the person and the data used to operate the device are together. basically FTC is like, no, it's not. They're different because you package up that data, you sell it to somebody else. so you can't say, well, we need this to operate the device, but then you turn around and you package it up and sell it to somebody else.

Leonard Lee: 27:47

Yeah.

Debbie Reynolds: 27:47

I think that separation is very. Critical because that has been one of the sticking points where you're like, oh, like if they're like, oh, Leonard, he doesn't need to know like all the, the safety X, Y, Z and the gobby good. That comes out of a car. It's like, well, no. If it is, for, maintenance or service or safety,

Leonard Lee: 28:08

yeah.

Debbie Reynolds: 28:09

You probably don't need to know that. But if that data's taking your phone book outta your phone when you collect, connect the Bluetooth and they're selling it to people, that's a problem.

Leonard Lee: 28:19

yeah,

Debbie Reynolds: 28:20

I'm glad that they're talking about that.

Leonard Lee: 28:22

Yeah. And a lot of this is rooted in the iot,

Debbie Reynolds: 28:26

that's right.

Leonard Lee: 28:26

Consumer iot, right? Yep. And a lot of people think that, ah, you know, the iot. That hype cycle, you know, went and stayed, went. It never went. It, it is always been there. and probably causing more problems.

Debbie Reynolds: 28:43

Yeah. Well, most

Leonard Lee: 28:44

people than solving, yeah.

Debbie Reynolds: 28:46

most people don't know. And most companies don't realize that they have more ILT devices, then they have computers.

Leonard Lee: 28:53

Yeah.

Debbie Reynolds: 28:54

So it's been a ubiquitous thing. people, go on Amazon and buy something, they connect it. Yeah. And as long as it does what they want it to do, it's fine. But they're not seeing that underneath layer. They don't know what's happening with data. And then, the device can change'cause you're doing Yeah. Uh, over the air updates and stuff like that. Yeah. So, I think that's a big issue.

Leonard Lee: 29:15

Yeah, and I think going back to one of our big themes for our podcast is the topic of trust. what's the outlook there? we're talking about a lot of these security and privacy related issues, but what do you think is the big theme this year for trust?

Debbie Reynolds: 29:32

Yeah. I guess I have two themes maybe we

Leonard Lee: 29:36

talked about.

Debbie Reynolds: 29:37

Yeah, we probably, it's

Leonard Lee: 29:38

an

Debbie Reynolds: 29:38

update every year. It's an update. It's updated, right, exactly. So the, the one thing is around trust for companies. I think they need to know how to. Collect or retain less, let's say you collect a lot of stuff. Mm-hmm. But then you only use it for its purpose and then once the purpose expires, it goes Bye-bye. Right?

Leonard Lee: 30:04

Yeah.

Debbie Reynolds: 30:05

Yeah. As opposed to collecting everything else. So I think that is one. For people to give less. So really be thoughtful About how you use data or who you give data to. Like, I'll give you an example. many years ago, I think Amazon was. Saying had told people, we'll give you a$10 coupon if you give us your, biometrics or something like that. And I'm like, that's not a good deal. That's not a fair exchange. You know what I mean? Maybe a hundred dollars, I'll think about it. But yeah. Or a free,

Leonard Lee: 30:38

free subscription to, prime video.

Debbie Reynolds: 30:40

So I think for the user, they need to think about is this. You know, it isn't the, it is never gonna be an even exchange because a relationship is asymetrical by nature.

Leonard Lee: 30:51

Yeah.

Debbie Reynolds: 30:51

But it shouldn't be just, just a bonkers, like you get nothing. You know, you give us everything, you get nothing. It shouldn't be like that.

Leonard Lee: 30:59

one of the big challenges, obviously, are the deep fakes.

Debbie Reynolds: 31:03

Yeah.

Leonard Lee: 31:03

And we talked about this ages ago.

Debbie Reynolds: 31:06

Totally.

Leonard Lee: 31:06

It was like a anti deepfake porn act or something like that. Now, apparently, and you know, I don't track all this stuff. I just assume it's happening because it's just probably the first thing that, you know, most people who get this technology tried to do with it is, yeah. there's, a legislation, right?

Debbie Reynolds: 31:24

There are a couple different ones. So a lot of these are state level. Yeah. Um, and what they've done, I think California's the first state. They have Revenge porn laws. And so some states are going back and they're revising those laws to add in non-consensual DeepFakes. Before it was like, okay, someone had a video or a picture of you and they put on the internet. That's what those laws are for. But they're like, well, this was not me, but someone made a like this of me. Yeah. And they're trying to taunt me or they're trying to Yeah. You know, do something harm me and something

Leonard Lee: 31:58

mine. Yeah.

Debbie Reynolds: 31:58

And so that part of the law hadn't been developed. And so we're seeing different states, like I think they updated their law and then there are a couple of other states are either trying to augment. those laws to add that in there, or they're trying to create new laws. But this is really a problem because this is an example, and I think we talked about it for years where. there will be things that happen in this age of AI For what? There is no adequate redress, right?

Leonard Lee: 32:27

yeah.

Debbie Reynolds: 32:27

Like a teenager. Someone creates like a deep fake video of this person. Like they feel like their life is over, their life is ruined or whatever. And like no law or red legislation is gonna undo. That damage. That harm, right? Yeah. That's what I'm really concerned about and that's one of the reasons why I got involved with like XR and stuff.'cause I'm like, things can happen here that are so super bad. Good.

Leonard Lee: 32:54

Yeah.

Debbie Reynolds: 32:54

And so I'm like, I'm all for prevention. I feel like the best way to solve A problem. So I'd rather be on the prevention side as opposed to, oh my God, this bad thing happened. Like, what happened to Johnny? Johnny's law, I don't like that, that, you know, some bad thing has happened to someone before we pass a law, because a law isn't a shield. a law does not stop things from happening. So that's what I want. I want to stop things from happening as opposed to clean up on aisle nine, something like that.

Leonard Lee: 33:23

And it makes you wonder, what are some of these guys training their models on?

Debbie Reynolds: 33:28

Yeah.

Leonard Lee: 33:29

You know, it, doesn't just know how to do this stuff.

Debbie Reynolds: 33:31

Right.

Leonard Lee: 33:32

It's like, okay guys, where you getting the content?

Debbie Reynolds: 33:36

Right.

Leonard Lee: 33:37

You know?

Debbie Reynolds: 33:37

right,

Leonard Lee: 33:37

yeah.

Debbie Reynolds: 33:39

Oh, yeah. Right. Capability doesn't, it's not natural, right. So it doesn't, it's not naturally occurring. It doesn't know. Right. So someone has to develop it to do such thing, so it just doesn't do

Leonard Lee: 33:53

Yeah.

Debbie Reynolds: 33:53

Stuff like this on its own. So you're right,

Leonard Lee: 33:55

otherwise you get the videos with the weird physics of people Right. Diving and, underwater ballet and stuff like that, where the physics are totally wrong, and it actually looks quite grotesque. here's the other thing. It's at some point you don't trust anything because it is all, all fake. Oh,

Debbie Reynolds: 34:12

okay.

Leonard Lee: 34:12

And then we all get, become desensitized to it. And so this is maybe a warning for an episode that we'll revisit. Maybe five years from now when this stuff actually becomes true. But this is gonna be a big societal problem. And that's the issue you mentioned, the data leakage issues or the unconsented sharing issues, all those things culminate into. malicious actors, being able to do a lot of really bad stuff with all of this available data. And one of the things that really shocked me, I just wanna share this with you, and if you, it's not on your radar, I wanna put it on your radar, Debbie is, there is, there's a thing called, I think it's called scam, GPT or something to that effect, which is an LLM that sits on top of the dark web and it allows threat actors to basically compose, attacks. I mean, this chat bot will help you construct targeted attacks based on information that's available in a dark web. And of course, we know everyone's social security number is on there now. And so think about these really precision attacks that can be generated now, and it, and this is because of all of this oversharing, but. Also the data breaches. All these data breaches that have happened. And so as we look at future data breaches and we. Become desensitized to these events. We do at some point have to recognize that it's these breaches that have fed the beast. You know, and I think now I, I'm gonna be really interested in attending RSAC conference this year, because I want to see what the tone is. It has changed over time from. Excitement. Hope to dread to holy. Oh my God.

Debbie Reynolds: 36:11

I know,

Leonard Lee: 36:11

this stuff is real and what's really kind of disturbing for me is that the level of awareness of what's coming is very low.

Debbie Reynolds: 36:20

Well, yeah, I shake my head when I see the reports coming out. Still the most popular passwords are 1, 2, 3, 4, 5. Or someone still thinks that they have a Nigerian uncle that has an inheritance. people are still falling for the old stuff. And so we're bringing technology that's way surpassed that. that is very concerning. even though these things are becoming more sophisticated, one pet peeve I used to have is when people say, you'll get a message from a scammer and the spelling is wrong. It's like they use, get Grammarly and chat GPT like everybody else. Yeah. That was a, that was a bad, yeah, that hasn't been true for a long, long time.

Leonard Lee: 37:00

Yeah, yeah, yeah, yeah. It's like dramatically correct.

Debbie Reynolds: 37:03

Scam

Leonard Lee: 37:04

artist.

Debbie Reynolds: 37:04

Totally right. They can, they'll pull a logo off of, you know, PayPal and make a email that looks like them. They're really good at that, right? But I think the thing from a person perspective or a company perspective, a lot of these are really just souped up. social engineering attacks. Because they still go back to trying to get you, you know, contact you out of the blue. It's some type of emergency thing that they contact you about and they want you to take action. So if you resist those three

Leonard Lee: 37:37

Yeah.

Debbie Reynolds: 37:37

Then you will be fine. Yeah. Right. Uh, so that's the, those are the things.

Leonard Lee: 37:42

I gotta share something with you. I recently received an email from somebody who, was thinking someone in my network for the introduction. This is a person I don't even know, emailing me. They CC'd a fake account of, a person that I do know in my network, someone that I know, pretty well and respect, and this, what is apparently a generated, marketing email. Eventually this will become phishing and it'll be used by bad actors, right? They're using your, your, network graph to now create more sophisticated, solicitations, which will eventually become much more sophisticated phishing attacks. So this is where we are, and it's like they say it'll only get better and for the consumer. It's gonna get worse. It gets better for the threat actor because this is a great tool for them. And this is one thing that I think. Is highly disregarded in Yeah, the responsible, ethical, and safe AI conversation, which is almost non-existent, right?

Debbie Reynolds: 39:05

Well, totally right, because it's like, oh, we could do all these great things. To me, technology is like a sword that cuts both ways. So yeah, I can do great things, but also people can use them in bad ways. The money that we're seeing being pumped into this industry just means that the tax will get worse and they'll come faster. The innovation comes faster.'cause the money is there. So we're seeing, the things the chat bots and different AI tools could do Six months ago, they could do different things now.

Leonard Lee: 39:35

Yeah.

Debbie Reynolds: 39:35

So, and to me that spills into iot as well, right. They have a sitting thing Absolutely. In your house waiting to take over the world.

Leonard Lee: 39:44

Yeah. Well, this has been a great, catch up. It's been way too long. Debbie, thank you so much for jumping on and sharing, your thoughts on, the outlook for 2026. I think there's a lot of accountability, responsibility, and focus on safety that needs to happen, I know it's a tough thing, but, I really appreciate the good work you do. it's rare. your lineage, your legacy in all of this is. So highly valuable, and I want the next curve audience to understand this because, you and I, always talk about safe ai, ethical ai, totally responsible ai. I don't think there's anything more distressing, year after year for us to witness that not happening, or at least being swept under the. table in the name of innovation, which is,

40:35

yeah,

Debbie Reynolds: 40:36

in

Leonard Lee: 40:36

my mind, dangerous innovation is not valuable. No innovation, unsafe Innovation is not, um,

Debbie Reynolds: 40:44

no, no, no.

Leonard Lee: 40:45

Innovation. So,

Debbie Reynolds: 40:46

no, no.

Leonard Lee: 40:47

Yeah.

Debbie Reynolds: 40:47

Well, I actually had the commissioner, privacy commissioner from New Zealand on my show recently, and she said something really interesting and this is, you know, I love cars. I like to use car analogies a lot because it really helps, right? but she was like, if breaks on cars didn't exist, we wouldn't be able to go fast. And I said, that's

Leonard Lee: 41:11

great. Yeah. That

Debbie Reynolds: 41:13

is so true.

Leonard Lee: 41:15

Yeah.

Debbie Reynolds: 41:15

Right. And so this is what we're talking about with safe innovation where, we still be like, maybe, have golf carts that are going like five miles an hour. Yeah. And we didn't have break. So

Leonard Lee: 41:27

that's awesome.

Debbie Reynolds: 41:28

So thinking about AI in that way, or emerging tech in that way,

Leonard Lee: 41:32

that's amazing.

Debbie Reynolds: 41:34

That's what allows us to go fast.

Leonard Lee: 41:35

Fast.

Debbie Reynolds: 41:36

Amazing.

Leonard Lee: 41:36

Yes. Yes. And with that, hey audience, everyone listening in, make sure you reach out to Debbie at www.debbiereynoldsconsulting.com. She's also on LinkedIn. She has her own podcast. it's, just phenomenal, education. On a constant basis if you follow her. So please do that. And then, please subscribe to our podcast and it'll be featured on the next Curve YouTube channel. check out the audio version on Buzz Sprout or find us on your favorite, podcast platform. And also subscribe to the next curve research portal@www.next curve.com. And our substack for the tech and industry insights that matter. Debbie, thank you so much and thank you for everything that you do.

Debbie Reynolds: 42:21

thank you. You've been such a great friend and collaborator. So it was always, I always get excited when I hear from you, so it's great.

Leonard Lee: 42:29

Yeah. And incre and congratulations on a million. That's awesome.

Debbie Reynolds: 42:33

Oh, thank you. Thank you. And congratulations. I love ILT Coffee Talk, by the way. Love this show.

Leonard Lee: 42:38

It's the best.

Debbie Reynolds: 42:39

It is.

Leonard Lee: 42:41

We'll see you

Debbie Reynolds: 42:42

All right.

Leonard Lee: 42:42

Friday

Debbie Reynolds: 42:43

later. Okay. Yes. Bye.