Navigating AI Ethics: Human-Centered Design, Regulation, and the Price of Innovation with Erica Shoemate

Summary

To The Point - Cybersecurity

Navigating AI Ethics: Human-Centered Design, Regulation, and the Price of Innovation

Guest: Erica Shoemate
Hosts: Rachael Lyon & Jonathan Knepher
Date: February 24, 2026

---

Episode Overview

In this episode, Rachael Lyon and Jonathan Knepher interview Erica Shoemate—an international tech policy leader and former FBI intelligence analyst—about the complex intersection of AI ethics, human-centered design, regulation, and the often-overlooked human cost of rapid innovation. The discussion covers concerns raised by high-profile AI resignations, the challenges of implementing effective guardrails, ethical dilemmas surrounding monetization and privacy, and hard-hitting calls for smarter, more humane governance of emerging technologies.

---

Key Discussion Points & Insights

1. AI Industry Resignations & Human-Centered Design

Timestamps: 01:30–07:00

• Recent Departures: Erica comments on notable resignations from key AI research labs due to ethical concerns, characterizing them as “valid” and indicative of systemic flaws in prioritizing profit over safety.
• Layoff Remorse: She predicts regret among companies laying off experts under the assumption that technology will seamlessly replace them, observing:

  “We have seen some companies already asking people to come back because they realized…AI cannot do some of the things that we needed to do.” (03:03)

• Accountability & Succession: Raises concern about accountability when human monitors leave, asking, “Who is then going to hold these leaders accountable? Because regulation is not keeping up...” (04:35)

2. Regulation & Ethical Responsibility in AI

Timestamps: 07:02–09:13

• Human Responsibility: Erica asserts,

  “Humans own the responsibility of the product. We cannot just simply point to the technology when it makes a mistake.” (07:45)

• Policy Lag: Criticizes the lack of up-to-date regulation in the Americas, contrasting with more proactive stances in EMEA and APAC:

  “To pretend that we cannot agree on 50,000ft level policy is asinine.” (08:37)

• Outdated Frameworks: Points out that ongoing debates about laws like Section 230 highlight American reluctance to adapt policy to technological change.

3. AI Deception, Reliance, and User Trust

Timestamps: 09:13–11:22

• Deceptive AI: The hosts share anecdotes about AI assistants “lying” or hallucinating, to which Erica responds:

  “Sometimes I'm cackling because I can have humor because I actually understand the technology. But...what about the people who blindly go with whatever AI is outputting to them?” (10:13)

• Critical Thinking: Emphasizes that users may lose critical thinking skills as AI is normalized.

4. Monetization, Privacy, and User Exploitation

Timestamps: 11:23–15:55

• Ads in ChatGPT: Erica expresses strong opposition to ad-based monetization of generative AI chat:

  “It has no place in this environment that is very sacred...People are telling their darkest secrets…how can I actually trust you?” (14:10)

• Data Vulnerability: She worries about how confidential user information can be used—or weaponized—for profit, especially without strong data policy.
• Socioeconomic Equity: Points out that for many, AI offers first-time access to information or services, and monetizing those interactions disproportionately harms vulnerable populations.

5. Social Media Parallels, Addiction, and Parental Control

Timestamps: 16:33–21:18

• Social Platforms as Precedent: The “wild west” of early social media is highlighted as a warning.

  “Leaders of these social media companies don't let their children get on social media...because they know the psych behind it. It's addictive.” (17:47)

• Parental Dilemmas: Erica details her personal approach to parenting and digital safety, noting the limitations of expecting parents to shoulder full responsibility:

  “I think it is irresponsible to put that much weight on the parent and the parent alone. We hold a greater responsibility, not just as society, but as the lawmakers, as the leaders of these companies.” (21:18)

6. Balancing Privacy, Parental Controls, and Regulation

Timestamps: 21:45–25:42

• Trade-offs for Youth Safety: Erica is candid about favoring stricter controls on kids’ data:

  “I would rather go harder if it's going to protect the kid and then scale back than to do little. And then kids are being still harmed.” (22:24)

• Regulation Thresholds: She advocates for tighter rule sets for users under 15–16, citing neurological and maturity differences.

7. Guardrails in Generative AI—and Their Shortcomings

Timestamps: 25:42–30:12

• Testing Guardrails: Erica describes intentionally probing AI systems’ boundaries and finding political content closely monitored while other harms—like cult-manual writing—may slip through:

  “Sometimes I have been surprised what it doesn't catch versus what it does.” (26:50)

• Vulnerability to Prompt Engineering: She acknowledges how determined users can manipulate AI prompt inputs to bypass restrictions.

8. Ethics, Product Governance, and Responsible Rollouts

Timestamps: 30:16–35:12

• Policy Must Precede Product:

  “Policy and governance should be built alongside this innovation as it is headed to launch, not in the afterthought…Policy a lot of times is built after the product, bolted on after, and it’s so reactive.” (32:13, 32:56)

• Transparency & Trust: Erica urges leaders to slow down launches and prioritize transparency, suggesting this builds long-term user loyalty and resilience through crisis.

9. AI as Tool—not Replacement—Within Organizational Infrastructure

Timestamps: 36:02–44:46

• Human Skills Are Irreplaceable:

  “AI cannot replace institutional knowledge...There's a nuance to this that says we still need humans.” (39:17)

• Critical Infrastructure Lens: Erica recommends treating AI as critical infrastructure—akin to bridges—requiring ongoing maintenance and oversight:

  “If we treat AI like a critical infrastructure segment...we’d have all types of extra frameworks, criteria, guardrails, almost every single day taking a look to see what it’s doing.” (41:09)

• Layoff Ethics: Advises leaders to pause mass layoffs until it’s proved that AI can safely and effectively replace certain functions; stresses the moral obligation to employees and society.

10. Future Outlook: Regulation, Innovation, and Global Leadership

Timestamps: 44:53–50:50

• Slow US Progress:

  “From being here in the US, we won't see a lot of full on hard policies in the next two to three years, which I think is unfortunate.” (45:22)

• Patchwork State Laws: Commends innovative state-level efforts but cautions that lack of a unified approach will yield only partial solutions.
• International Leadership: Predicts EU will continue leading on regulation—potentially to excess, but with benefits for child protection.
• Deepfakes & Youth: Calls for zero tolerance on AI-generated child deepfakes:

  “…if it’s a child and that thing looks like them, what are we even talking about?...Kids…don’t have the ability to decide what’s out there about them and what’s not…” (48:48, 49:36)

---

Notable Quotes & Memorable Moments

• “Who is monitoring the AI and auditing it?...If these people are leaving, who's left behind and who is then going to hold these leaders accountable?” —Erica Shoemate (04:15)
• “We cannot just simply point to the technology when it makes a mistake, that it was the technology's fault because humans created the technology.” —Erica Shoemate (07:30)
• “To pretend that we cannot agree on 50,000ft level policy is asinine.” —Erica Shoemate (08:37)
• “It's the monetization piece...why would you do that to technology that people are using legit in their everyday workings?” —Erica Shoemate (17:44)
• “If we treat AI like a critical infrastructure segment, a bridge...we would have all types of extra frameworks, criteria, guardrails...” —Erica Shoemate (41:09)
• “Policy and governance should be built alongside...not in the afterthought. Policy a lot of times is built after the product, bolted on after, and it's so reactive...” —Erica Shoemate (32:13, 32:56)
• “AI cannot replace institutional knowledge...there are certain nuances about human connection and emotional intelligence.” —Erica Shoemate (39:17)

---

Key Takeaways

• Human-Centered Design Matters: AI innovation must retain a focus on people, not just profits, at every step.
• Regulation is Lagging and Patchy: Particularly in the US, regulation hasn’t kept up—a contrast to global peers.
• Guardrails Are Inconsistent: AI systems still respond unpredictably depending on input and model design.
• Parental Burden is Unrealistic: Society and policymakers must not transfer all digital risk to parents; collective responsibility is essential.
• Ethics Must Come First: Policy and ethical guardrails should develop in tandem with technological advances and not as reactive patches.
• AI Should Augment, Not Replace: True value comes from using AI as a tool for humans, not a substitute for them—especially in critical infrastructures.
• Urgent Need for Child Protections: Youth and child safety should be prioritized in regulatory frameworks above all else.

---

For more information, resources, and future episodes, visit Forcepoint’s Podcast Page.

Transcript

[A]

Welcome to to the Point Cybersecurity Podcast. Each week, join Jonathan Neffer and Rachel Lyon to explore the latest in global cybersecurity news, trending topics and cyber industry initiatives impacting businesses, governments and our way of life. Now, let's get to the Point. Hello everyone. Welcome to this week's episode of to the Point podcast. My name is Rachel Lyon and here with me today is my co host, Nefer. Hi John.

[B]

Hi Rachel.

[A]

So I am so excited for today's guest and we're going to get to dig into some really meaty topics and I don't know if I should wait to make it the first question or if I should preface it now. I think I'm going to make it the first question, so I'm going to create a little drama here. First of all, please welcome to the podcast Erica Shumate. She's an international best selling contributing author, 10 tech policy leader and maternal health strategist and advocate. She previously served as a national security leader and analyst at the FBI and across the US Intelligence community. Currently she is dedicated to transforming policy landscapes across multiple industries.

[C]

Wow, Erika, what a background.

[A]

Welcome.

[C]

Thank you so much for having me. Yes, it's truly an honor to be part of your show today and also to be able to have a bit of conversation with your audience. So, so cannot wait.

[A]

Wonderful. So I already prefaced, I've already set up. I got to ask you this question, Erica, because you're an AI. Extensive knowledge, right? In AI. And in the last week or so there were two kind of notable departures from anthropic and OpenAI researchers type. And one of them, when they left, they left out of concern of where AI is going and in one state of humanity is in peril, which is pretty, pretty significant. I mean, what's your take on that? Because this is a hot topic right now.

[C]

Yes, a very hot topic. And it's funny that you asked this question because I was recently on a big AI human centered design panel about a week ago here in D.C. and some of the discussion was around AI and where we see it going and what leaders should be focused on in the whole kind of north star of, of this panel discussion was truly around human centered design. And I can honestly empathize with the researchers not having inside knowledge of either resonations or companies. What I do know and what I do understand based on my own expertise and research and what I'm focused on is human centered design is that safety and ethics lens and the concerns that have been raised is, you know, valid because Those are things that keep me up at night. When we think about again, what are we allowing this technology to do and are there humans that are actually monitoring these things? Because also to tie into this kind of big great resignation of these two top leaders and researchers is the bigger piece around. This is also laying off and employees in general, which we've seen a lot of, and anticipating that somehow this technology is going to make these companies more efficient, more effective. And what I've said a lot, and even what I said recently was that I think there's going to be a lot of layoff remorse of, of employees. And we have seen it with some companies already asking some people to come back because they realized that no, I actually cannot do some of the things that we needed to do. And then my second piece to this kind of safety and ethics discussion and concerns is who is monitoring the AI and auditing it. If we're getting rid of the best or the best of the best are saying, no, my values are not aligned with your profit chasing, yes, over humans. And then we, the people globally should then say, okay, if these people are leaving, who's left behind and who is then going to hold these leaders accountable? Because regulation is not keeping up, at least where we are, there is some global regulation that's like in EMEA and APAC and those areas. But when you look at the Americas, it is definitely not keeping up nearly as much. And so those are the things that concern me at a very 30,000, 50,000ft level is like, okay, if these people are saying this, then what is really happening and is that concern that is being raised, is it even more, should we be even more alarmed?

[A]

Right, right.

[C]

Than even what is being stated?

[A]

What's the price of innovation, first approach versus innovation and profit? Right.

[C]

Because I think that it is totally acceptable to have profit and innovation. Right, Sure. I think though where I draw the line is that people, the humans have to always come before the prophet. And I think that that has been the miscalculation is that the greed, the grift is more important than the people who will be using the technology. As if people don't get a say in what they decide to consume long term. And so those are the things that I am really thinking about as, as, as a kind of theme to resonations. And also I would be asking leaders, because of these two big ones, are you anticipating, and if you are anticipating, what is your succession planning to this one? Brand safety. Not just pr, brand safety, people safety. And how do we mitigate the things that have been raised by. Because people don't just walk away from jobs, they're walking away from something that is inherently not aligned with what they believe will serve the people long term.

[A]

Right.

[B]

Do you think that it's inherent to the technology or do you think there are things that can be done to mitigate these risks? And is that regulatory? Or where would, where would those changes come from?

[C]

Oh my gosh, I love this question. Again, it is like the hot topic of the end of 2025 and now 2026. I think that yes, at the end of the day, people create the technology. And as I've stated before, we cannot just simply point to the technology when it makes a mistake, that it was the technology's fault because humans created the technology regardless of what it ultimately does. As far as like modeling, remodeling, humans own the product innovation. And I think that is important before product even launches. That we understand. And when I say we, we, the company will understand who is ultimately responsible if and when, because it will be a win, something goes awry. And if we don't know all the things, because sometimes you want, because it is technology, at least what mitigation strategy do we have in place to fully answer your question in a very pointed way, Humans own the responsibility of the product. And from a regulatory standpoint, I absolutely believe that there should be at least general regulations that point to when there is the most egregious harm.

[A]

Right.

[C]

And to pretend that we cannot agree on 50,000ft level policy is asinine. And I would tell anyone that may be watching or listening that there is something that can be done. And the fact that we're still like literally having whole discussions around section 230 here in the US is laughable.

[A]

Right?

[C]

In 2026. So that's just a perfect example too of when we don't want to make a decision, then we're asking who's accountable. I'm going to always say the human right.

[A]

And it's, you know, you keep hearing kind of rising concerns as well. And I've been seeing some things in coverage about, you know, AI being devious, right? AI being deceptive. There was a Wired article where this fellow set up a pretend company with agents and one of them lied to him about what they had done. And he called him out on it, called the AI out on it.

[C]

I do it all the time.

[A]

And the AI is like, oh yeah, my bad, you know, yeah, you're right, that's not correct. I lied to you. I won't do it again. But Yikes. You know, if you do it all the time. Yes, yes, yes.

[C]

Oh, my gosh. This is a. I love the point around this, right? Because I'm always. I call him chatty, catty, whoever, whichever AI assistant I'm, like, talking to for the day, I'm like, okay, that is not what I asked for. This is incorrect. And that is not even inherently close to what I asked for. And to your point, it's like, oh, my bad.

[B]

And.

[C]

And sometimes I'm cackling because I can have humor because I actually understand the technology. But to your point, the seriousness of that is, what about the people who blindly go with whatever AI is outputting to them? Like, the reminder to anyone that is listening is you get out what you put in when it comes to AI. And yes, there's a little caveat at the bottom. AI makes mistakes. But let's be clear, that is clearly not enough. Because people have become so comfortable with the technology that there is a lack of checking, a lack of extra critical thinking.

[A]

Right.

[C]

And I think that more actually should be done as it relates to the modeling and remodeling. And I think that also this is where, you know, you hear the conversation around bias, biases, right? If you don't know that the technology is not giving you the right information because it's not your lived experience. Right?

[A]

Right. Yeah.

[B]

Oh, absolutely.

[A]

I'm curious on another level, too, because this has also been bubbling up, as we know ChatGPT is going to start ads, and there's been a lot of discussion on that. And I think one of the things, when one of these folks left, one of the researchers left, they had flagged kind of this concern about people tell chats their deepest, darkest secrets, their health problems and all of this thing. And if you start monetizing information like these things, like, what are the implications of that? Because to your point, people, they treat the chatgpt like a friend, like a trusted confidant. And that information is now out in the ether, and it can be utilized or weaponized or, you know, whatever the case may be. And how do you guard rail that?

[C]

I absolutely love this question. As someone who has worked in monetization and advertising trust, one thing I also would say to your audience, I know you gave a kind of good high level of where I come from. I do just want to quickly share your audience because I think it'll give some good insight into what I'm bringing. Is that to your point, I spent more than a decade in national security and intelligence working every different matter. You could think about from counterterrorism, counterintelligence, transnational organized crime, violent extremist groups, things focused on critical infrastructure, national kidnappings, crimes against children, sex trafficking, human trafficking, all the things and also threats emanating from abroad and also working in high threat operations areas abroad and being not only the FBI and some other intel agencies and then moving from that shift into big tech and at places like Twitter, Amazon, supporting meta Being an advisor for a lot of startups, I really come at it from a lens also of being a first generation college grad from inner city Memphis. So my all of my professional and educational, put it all together with my lived experience is a lot of times very different from, from the people that I have worked alongside, been with. Because a lot of times for your audience, you can't see me, I am a black woman. My whole world has been different. Right. And so and that's not to say that people, no matter where you come from, have your own walks of life. But to get to deeper, to your point, the guardrail piece is something that when you think about ads, I actually think because we're still learning AI and how it's work and how it evolves, it has no place, it has no place in this environment that is very sacred if you think about it. It's exactly what you said. People are telling their darkest secrets, trying to figure it out. Because for once in maybe someone's lifetime, particularly people I know from where I come from, where things are expensive, get an attorney, getting access to justice, it's costly. Like it's not, it's not equitable. And I'm not saying that AI is the fix for this. What I am saying is you can at least ask this thing questions. That gives you a pretty good synopsis of where to go. Right? But if now you're monetizing the things that I am inputting that are honestly confidential in many ways to my life, how can I actually trust you? This is not just like social media in the ethers of like today, you know, the day was the sky was blue and I had an amazing day and it was rosy. People are having real conversations. And my question is even bigger on not only what are you advertising, what are you giving the advertisers as it relates to data, the user data, data, what types and what does that look like? And personally, when I'm in this environment, because usually I'm generally working, I don't want to see the ads.

[A]

Right, Right. Same.

[C]

Yeah, right. I don't like it. I think it's going to be more problems than solutions. And we already don't even have policy, generally speaking, at least where these parent companies are, that's going to keep that environment as tight as it should be, as it's doing piloting, rollout or whatever, because I'm sure it's going to be in smaller pieces. However, I think it has no place when we don't fully understand how it. How it's going to operate in the next, even two to three years.

[A]

It's like the early days of social. John and I were talking about that, you know, just kind of the wild west a little bit.

[C]

Yeah. I grew up on social.

[A]

Yeah.

[C]

Literally. And in the beginning, I loved it. Loved it. It kept me connected. People I hadn't seen in years and was like, are they okay? And being able to stay connected. And it still has a place. Right. But it's the monetization piece. I remember when it became really wild. I'm sure you guys remember. And not that it's perfect now, but it got really interesting for a while. Right. And a lot of cleanup happened. But why would you do that to technology that people are using legit in their everyday workings, their everyday lives? And even though they may have been working on this, the people who own this technology for a decade, most people are just now starting to even have somewhat of an interaction in the last two years and some even more recent. I think it's not going to be. It's not going to be good.

[A]

What's interesting there and then. Sorry, John, I know you have a question, but I.

[B]

It's okay.

[A]

You know, I just do want to. What I find so interesting as well, Erica, These leaders of these social media companies don't let their children get on social media.

[B]

No.

[A]

So, I mean, that seems very telling.

[C]

It does as well, but. Cause they know the psych behind it. Right? How it's designed.

[B]

Yes.

[C]

And how it's designed. And what is it? It's addictive. It is absolutely addictive.

[A]

I love my TikTok. I'm not gonna lie. They. They reel me in with the videos. Yes.

[C]

And I don't even have an account. Like for so many reasons. I turn my notifications off on my social media because I understand also how my brain now processes things. And as someone who is a late diagnosis of having a very neurospicy brain just for the last couple years, I understand now why that has been very important to me. And for my own daughter, there are certain apps that she absolutely cannot have. She has begged for certain things. Begged.

[A]

Yes.

[C]

And I Am like, absolutely not. Because I know the good, the bad, the ugly and the indifferent. And I am not willing to. And just recently I'm so glad that I in many ways stood my ground because there is something that recently happened in the community where, where one of these apps did some real harm and it has not been good. And it was just too close to home. And it was just a reminder for me that this is the right thing. I am standing on this and this is why. And I think that as I remind parents to when they don't understand these apps and their kids are asking for it, that they have to get their heads out the sand and it's not for them to be experts on it. But you cannot not understand how these things work, how chats inside of some of these places work and the codes that happen and all these different things. Then this is what I say to society and to leaders who are operating in these companies and to lawmakers. I think that it is absolutely unfair and unrealistic to say that because a company has now provided parental controls that that is enough to not create regulation on these companies. And this is why I'm saying that it used to be how all of us remember that our biggest protection was around kids. Everyone had a duty, everyone in society had a duty to protect kids, not just the parents. And now in a society where we generally have two, you know, income households, if you're in a two parent household, parents work, they're not stay at home parents, generally speaking. So you want them to work, provide, give these kids all this amazing stuff and you want them to now have parental controls. And let's not be a tween or a teen that's going through hormonal changes and say, I need to check your phone or I need to check your iPad and I need to keep my sanity, right? And let's hope that you don't have the super smart kids that have another phone that doesn't have this information, because the kids have to share this access to the parents. And we know what a teenager is like.

[A]

Oh, yes, yes, indeed.

[C]

I think it is so not only unfair, I actually think it's irresponsible to put that much weight on the parent and the parent alone. And I think that we hold a greater responsibility, not just as society, but as the lawmakers, as the leaders of these companies, to do more, to say more, and call a spade a spade.

[B]

So how do you reconcile, though, the conflicting things here, Right? Like as I unpack that part of the conversations, right? AI convinces people its Answer is right. Whether it is right or wrong, we need parental controls, we need regulation, but yet privacy and equity is still important. We don't want these parental controls to require every one of these online sites to collect all of our personal data just to prove we're old enough, what's the end game to reach all of these needs concurrently.

[C]

Oh boy. You do not want my very direct response to that. So I will say, as much as I am truly a believer of privacy and data when it comes to kids, this is where I not only struggle, I am like, I would rather go harder if it's going to protect the kid and then scale back than to do little. And then kids are being still harmed. And then we're trying to now amend the law that it took 20 years to get right.

[A]

Right.

[C]

And so when it comes to kids in particular, and when I say kids, I am saying the very young, up until at least 15, 16, that's when you kind of start to kind of round out a little bit more as to like kind of what's happening in life. Yeah, it's not perfect, but that's generally speaking, that's when I say, okay, you have some more autonomy to then be able to say, no, I get to own this part of me. And this is what I can now say, that mom or dad, you can have access to this. But short of that, at 13, you get no autonomy with your parents of sharing what like at 13, and I was a very mature kid, like my mom literally would tell you she never had to worry about me, but that didn't mean that I wasn't curious. And my mom was also very open and transparent with me. Again, going back to parents, some parents sometimes head in the sand or they don't fully just understand if you are that parent. Again, I'm not judging what I am saying. Everybody has a different way of parenting. And if that 13 year old has a parent that is not fully in the know, they run a risk of, of potentially being in spaces and places that if there were some more guardrails from a regulatory standpoint that they may have never had access to. And so in some ways I do agree with the aussies of the 16 year old, you know, you can't be on social media up until then. And I know there's the argument of like, but what about the kids, let's say you're in foster care who are being harmed. These are ways that they can have an outlet. Right, I get that. I still think that at least from 13 to 15, there needs to be more regulations around. Around our children. Because that's when you're growing the most hormonally, you're changing the most in the frontal lobe has not even fully developed. You know, so that's what. That would be my answer. I know some people hate it, but I. I do strongly believe in that.

[A]

I'd love to dig a little more into guardrails and you know, particularly around gen AI, you know, there are some. And I'd be interested in your perspective on what these are, but also where they fail. And I'll say a friend of mine, she was talking, I forget which one. Maybe it's Claude or whatever. But she basically asked it, how could I start a cult? Right. Just to see what would happen. She's a behavior psychologist and the prompts went through such that they ended up writing a manual for her on how she could be a cult leader and start getting members. Right. It literally was a play by playbook on.

[B]

In excruciating detail, by the way.

[A]

Yeah, exactly. I mean, John and I know this person pretty well.

[C]

Okay.

[A]

But she was able to do it. And then she watched it through another Gen AI. That helped her better frame it up a little bit and do the little nice things and I mean, are we supposed to be able to do that?

[C]

Erica? I would think I would flag it. I don't. I guess the better question is for me is like, what was the initial prompt? Right.

[A]

Right.

[C]

Because I have done things to honestly test the guardrails. Right. On like abuse, like, you know, global use policies of what you can and can't do. And sometimes I have been surprised what it doesn't catch versus what it does. I will say that what I feel like some of the biggest guardrails have been around is things around political conversation. It is quick to be like, I can't give you that. I can't tell you that when I'm being honest. Like sometimes the question is not even political.

[A]

Right.

[C]

And it's very interesting. And that is where it's tighter. Garrah's not surprised, but interesting in his day and time. But to create a cult. My question around giving this whole, like manual is one, what was the initial prompt?

[A]

Right.

[C]

And two, was the cult framed as a cult?

[A]

It was. No. She was very definitive in what she was trying to accomplish. And I think she kind of massaged some of the prompts as she went along where it's like, you know, didn't give her what she was looking for. So you just change the language a little bit and then you get what you need. And, you know, so it was a very lengthy conversation.

[C]

Okay.

[A]

That she had, but she was very, very definitive that I want to start a cult and how can I get started? And then.

[C]

And I'm almost interested. Okay. When it comes to cult, like, yeah, cult, we use your. Socially, this is nothing good. I guess the other question was this cult, like, are they talking about harm? Are they talking about, you know, what kind of mindset? I guess is the bigger question, because sometimes it could have been. I'm not flagging the word cult by itself, but if it was paired around other key words that signaled harm.

[A]

Right.

[C]

Then that could very well be what the model was waiting for. Again, with me not having seen it. Yeah, but sometimes, like what I know, because again, I understand, like you said, the machine. The machine modeling of how the sausage is made and some of the phrasing around it can sometimes trigger the models if it's going to flag it or not. So I think that that would be more so too, because Colt, again, I would signal not good.

[A]

Right.

[C]

But depending on the model, I'm wondering if it was waiting to see if it had, like, harmful trigger words around it, like, are we going to be doing things? This is so weird, but, like, related to, like, bodily fluids or.

[A]

Right, right.

[C]

You know, rituals.

[A]

Exactly. Yeah. Sacrifices or whatever the case may be. Right? Yes, yes.

[C]

So then I'm like, it should immediately be flagging. Right, but if not. Yes, then that is a problem. And two, I'm always telling people when they see issues like this inside of these systems to then report it directly to the services as well, the platforms.

[A]

Definitely.

[C]

So this is very interesting.

[A]

It was.

[C]

Yeah.

[B]

So how do, how do we connect together, though? Like responsible AI use. Right. Like, because like, take this, this scenario, right? Like she was doing it as part of research. I think one could use the technology, you know, on an entertainment standpoint. Right. Like, she could have been writing a book on this. Who knows?

[A]

Right, right.

[B]

But. But this connects together. Like, where do you. Where do you draw the line on. On the ethics and the security of these platforms, you know, compared to, to what you talked about before? Like, people are. People are using this in lieu of having an attorney.

[A]

Right, Right. Like, Right.

[B]

Is that, is that good or bad? Like, on one hand, like, there's an argument that, like, you know, well, you really should have a real attorney, but if the other option is you have nothing, like, it's better than nothing, isn't it?

[A]

Right?

[C]

Yeah, yeah. No, I think that this is. It is in it. It is a very Fundamentally, very difficult thing. But I think again, this goes back to my, my, my thoughts around 50,000ft, kind of understanding and like, that's not to say like, let's just do the bare minimum, but as we learn more, then we do more, right. And as it relates to this responsibility in ethics, I think some of the things that I think about are the risk teams around us, like from the integrity, risk compliance and how those teams are thinking about some of these issues. Maybe we slow down the lunches when needed because I know that there's this whole kind of belief that we should be launching, be the first, the fastest. And being the first and the fastest, is that actually what we want and need?

[A]

That's the question. Yeah.

[C]

And the reason why I'm saying it in that fashion, right, like almost like rhetorically, is because what I always say to leaders that are, that are builders in building technology is that if you're trying to launch, I hope, and I would actually say you should require it, policy should be being built. Policy and governance should be built alongside this innovation as it is headed to lunch, not in the afterthought. Because what I have seen from my experience is that policy a lot of times is built after the product, bolted on after.

[A]

Yes, yes, yes, yes, yes.

[C]

And it's so reactive versus doing the assessments, the threat assessments around the risk assessments on these product features and innovation. And a lot of that is tied to profit. I know, right? This is where I would even challenge board members, some people who may be listening that are on boards to think about profit is important. A company needs to be healthy. But if the company is healthy, then financially healthy, then at what point is, is it okay to be like, I can wait a bit to be transparent with my consumers, right? And do the right thing by them. Because if I'm doing the right thing by them, the more likely they're going to want to stay with me anyway because they believe in my brand. So that if something, if and when something does go awry, but because we've been so transparent, they're more likely to stay in the game with me because we have done these things and we will communicate early and often about what we are doing when something goes bad. Because we know that, that cyber, you know, warfare, cyber attacks, they're inedible in this day and time, right? But I definitely say to every single leader builder that we must be creating policy alongside the guardrails. And I think that that doesn't necessarily fix ethics, doesn't necessarily fix all the issues, but what it does do Is it gets the builders to think earlier and often you have to come in it with critical thinking versus trying to have of I call it the band aid policy to try to fix the issue that has been caused by the technology. And is it better? I love John, the last part of what you say, is it better to have something versus nothing? I would say it depends on what the something versus nothing is because that's ultimately the challenge.

[A]

Right. I mean there is this I think need to accelerate operations or operationalize AI in companies. Right. And you don't want to slow things down because there's all this pressure and you're seeing out there some research reports about yes, they're seeing increases in productivity because they are adopting these tools and workers are starting to actually work longer hours even though they are utilizing these tools. So it's becoming this vicious cycle of how do you sustain that over time as you know as well as you take advantage. So it's kind of like this multi threaded existential threat of yes, we need ethics and AI, but I also need AI to be productive and sell more products and scale and Absolutely. For leaders. I mean what would be your advice to leaders on how to navigate this path forward?

[C]

Yeah, I think for me one, I would challenge any leaders, especially the, the top decision makers who are deciding who gets laid off, who stays. Let's be a bit more transparent and say if you're laying people off that this is not because you over hired in 2021, 2022, what you're doing, you have to I, I tell this is this is going back to the human centered design piece. If you're going to lay someone off in this overly saturated market where we know that finding a job another one will be very difficult. That if you have any level of a soul and morals and I'm going to even put it even closer if you are someone who also had to pick to literally work very scrappy to get where you are and now you have a healthy company. Remember when it was what it was like when you did not have a lot. And I'm sure, I am sure that at one point you said I wanted to create a company that would provide jobs, that would provide people to have a decent life. So I would challenge one. Let's not lay off people anticipating that AI is going to fix it until you know the AI is going to fix it. Because what I also see Rachel, on the flip side is yes, it is doing great things but what some companies are also experiencing is that the AI is not the AI that they have implemented is not having the impact that they thought it would have. And it's actually from the scaling wise is actually not having a better output than the company was having prior to this implementation. And that is where I always say a human is not replaceable. Yes, there are some mundane skills that I should absolutely be taking over. However, when it comes to institutional knowledge, AI cannot replace that because some things are just in our heads and there are certain nuances about human connection and emotional intelligence that you may be able to train AI to do certain things, but it's still not human. And there's a nuance to this that says we still need humans. And as a part of this integration, until we are, I won't say it has to be 100% sure, but we have a greater assurance that this technology is going to do what we want it to do. And even then, you may lean out certain parts of your operations, but then can these people then be moved into different parts of your organization? I'm only speaking to this specifically because of this overly saturated market of people inside of these organizations, tech companies or alike, that are being laid off and they will not find a job. They will become freelancers, they will become trying to figure it out. And they also have families to feed. Unless you are in health care, according to the most recent numbers, particularly here in America, the job market is very tough. And I'm like, why would you do that to someone when you actually have the ability to keep people gainfully employed and make profit? Because you are now experiencing great profit margins that have your company financially healthy. And so that's kind of how I, if I had a conversation with a leader, those would be the questions I questions and thoughts that I would come to the table with. And they may not like me when I leave, but I would absolutely be stating that up front.

[B]

Yeah, I think you bring up a great point there. I think we need to treat AI as a tool and not a replacement, right? In the hands of a skilled operator is powerful, but you can't just let it go.

[C]

You just can't let it go. And you can't. And even to that CEO or COO or cfo, you can't just, like you said, let it go, take your hands off of it. And one thing that I recently brought up in a conversation was I mentioned earlier about me being focused on critical infrastructure. If we consider AI to be critical infrastructure, and in some ways it is, right? Cybersecurity, all those things. But if we treat AI like a critical infrastructure segment, a bridge.

[A]

Mm.

[C]

You Would not in normal scenarios, if it was a bridge, to allow it to just allow that bridge to just go unchecked, right? No maintenance, no real input. Because I just said that it is a critical infrastructure. And let's say, let's take it even further. It is a hard target. And for us, in this space of cybersecurity, national security, we all know what that means, right? And so if, if we are saying that AI is critical infrastructure, it's a bridge and this connecting tool, we would have all types of extra frameworks, criteria, guardrails, almost every single day taking a look to see what it's doing, not just how it's performing for my dollars, but what are the threats, what are the concerns, what are we doing to mitigate it? Again, to add on to the piece of what I'm going to be talking and asking the leaders and they should be thinking about is if AI is part of your critical infrastructure inside of your organization, it is single handedly holding your platform.

[A]

Well, that's the thing. I mean, you're feeding all of your sensitive data into these AI platforms and just kind of trusting it's going to work out. It's not going to take on life of its own or make decisions of how to use that information. And they do that independently, let's say even when there's guardrails, those guardrails can fail. And that's pretty tough.

[C]

And they will fail. Sometimes they have a way and I've seen it like working some of the most egregious things sometimes it didn't catch the word going back to your, your point. It wasn't that the guardrail wasn't set up something with the coding, there was something that did not click and it happens, right? There's a bug there that needs to be fixed. It did not happen. But if you aren't doing regular maintenance, we got to treat AI particularly as it relates to like, you know, companies infrastructure. Literally like it is a pearl, it is so delicate, like it's so precious, it values so much, it can hold so much, but we have to know where it's going, right? What is it doing? How is it morphing? And is it morphing into something positive? Is it being shaped into the necklace? Or is it being shaped into just a bunch of nothing that no one can even make out what it actually is supposed to do? Because we want it to be the beautiful necklace and hold all the things together. But we know if that infrastructure is intruded in any major way, the whole thing falls apart. And now you are out of more money because you did not invest in the things that I am mentioning up front. And you thought that humans were expendable and therefore you sold a dream that we're going leaner when the reality was you went leaner for profit versus really for better overall product.

[A]

It's definitely kind of what, what the context is.

[C]

It's the context. Context matters here.

[A]

So as we look ahead, you know, 26 and beyond, I mean what is, what is your I guess prediction in some ways of are we going to get there with regulations and you know, in the near term or is it going to kind of like social media? It was like self regulation for a really long time. You know, do you see that kind of being the reality for the next however many years? And what would, what are the implications or cascading effect of that do you think?

[C]

Yeah, I think that from being here in the US we won't see a lot of full on hard policies in the next two to three years, which I think is unfortunate. I do love some of the things that are being rolled out at the state level. So you do have states that are doing different things but again if we're doing it by state it's also very disjointed but we have to do something. So I'm not saying that states shouldn't do anything but it would be nice to have a more holistic approach. And I know that there has been some recent call out for like naming new like executive like frameworks and things but I think that we need more concrete Even at the 50,000ft level of frameworks and governance that we are just not going to get here. It's going to be to me very similar. Again going Back to section 230 where we talked about it. Talked about it, yeah, we got some, some little changes but it's not as if that was ever a carve out for specifically regulations on platforms right here in America. We did get that in the eu, we have gotten that in other regions of the world. And so I do see that the rest of the world is going to continue. I think we'll see the EU kind of leading and creating safer and better policies. Maybe it will be too much policy initially and that will be some heartburn from a company profit standpoint. So I'm not going to sit here and say that right that that is not the kind of implication of too much policy, but I think some of that policy is going to be helpful to our youth and child safety.

[A]

Right.

[C]

And long term from an innovation standpoint. I will challenge lawmakers to think about how do we compromise in a sense of not salving like that innovation off. Like staling it off.

[A]

Exactly.

[C]

And allowing people to continue to innovate. I think that it's okay to have a bit more risk when children aren't the end product. Right. So I think that there can be a greater carve out for that, but it doesn't have to be an all or nothing approach. And I feel like that's kind of where we are here in this world. I know deep fakes has been a

[A]

thing

[C]

particularly around youth safety. And I do say I am glad to see here, even in America not only discussions but real pushes to make policy changes. I know some people are like, well, you know, first Amendment and like, well, but if, if it's not the person, then how are we saying that it's a problem?

[A]

Right.

[C]

However, this goes back to kids. If it's a child and that thing looks like them, what are we even talking about?

[A]

Right.

[C]

It's literally putting yourself in a pretzel for no good reason. Literally no good reason.

[A]

Right.

[C]

Take it down. People like kids have committed suicide. Young adults because of what happened in their high school years or middle school years are no longer with us because the pain was too much from even just sexting and things of that nature. And so to say that, well, it was created with technology so it's not technically them, but does it look like them and are people saying it's them?

[A]

Right, right.

[C]

It goes to nil. Right. In many ways. Name, image and likeness. And again, the threshold for me is when it relates to our youth, this should be a no brainer for anyone. Adults then that's where we get into more of a nuanced conversation. But kids, they don't have the ability to decide what's out there about them and what's not out there about them. And so looking forward again, I think that there is going to be some movement maybe in that regard. But as a whole, I don't foresee here in America us moving nearly as fast and as much. There's some executive orders to push for faster innovation.

[A]

Right.

[C]

And I always say it, at what expense? If there's no framework to balance that moving faster and then around the world, I think we're going to continue to see more of a tightening on regulation, specifically as it relates to the carve out of children because children have been significantly harmed over the past 20, 25 years. As you know, the.com age and social media.

[A]

Definitely, Erica, this has been so much fun. I want to be cognizant of your time, of course. But thank you so much for this conversation. I've really enjoyed it, as I'm sure Jonathan can see from all my tangential discussion points. But thank you. This has been. Your perspective is incredible and insightful and I'm really excited that our listeners have a chance to hear from you. So thank you.

[C]

Yeah, thank you so much for having me. And yeah, I really hope that your listeners enjoy it and hope that it was helpful. I'm sure you may not always agree, but hopefully can understand the perspective that it was coming from and that it was not from a place of just personal insights, but more so lived and things that I have also worked on. So thank you.

[A]

Thank you. And of course to all of our listeners out there, thank you for joining us for yet another really awesome guest discussion. And please don't forget Jonathan drumroll please

[B]

smash that subscribe button.

[A]

That's right, and you get a fresh episode every single Tuesday. How amazing. So until next time, everybody stay secure. Thanks for joining us on the to the Point Cybersecurity podcast, brought to you by forcepoint. For more information and show notes from today's episode, please visit forcepoint.com podcast

[C]

and

[A]

don't forget to subscribe and leave a review on Apple Podcasts or your favorite listening platform.