Welcome to the Bulletproof Your Marketplace podcast, where we break down the strategies, stories, and insights from my book, Bulletproof Your Marketplace, one chapter at a time. Whether you're a founder, legal counsel, or trust in safety leader, or anywhere in between, this series is built to help you navigate risk and scale with confidence. In each episode, we'll dig into a chapter of the book, explore the key themes, and give you actionable takeaways to help make your marketplace safer, stronger, and more sustainable. On this episode, I'm joined by Yikater, who has a wealth of experience working at some of the largest global platforms.
You're welcome. Thank you. So to kick things off, can you introduce yourself a little bit more, talk about your experience and history working with marketplaces in the various digital platforms? Yeah, sure.
So I worked as a risk and regulatory management professional in some of the largest online platforms. One was more user-generated, focused than the other. But I've also had experience with other online marketplaces, and both at Google and X, I had to align operations with strategic policy goals. And I had a risk and regulatory compliance, especially with DSA, but also with US child safety related laws.
My exposure to Amazon, I had a lot of discussions with its risk team. So I'm also familiar with some high profile cases there that relate to Section 230 and product liability, which we'll get into later, which maybe I can shed some light and insight on there as well. And before that, I was in the government managing defense and national security projects on more compliance and audit related focus. Amazing.
So you're the perfect person to talk about this. In this episode, we're going to talk about the chapter on Section 230. And at a really high level, can you talk about why Section 230 is so important for platforms right now? Yeah, sure.
As your chapter really nicely lays out, it's really a transformative piece of legislation that gives broad immunity to online marketplaces known as interactive computer users. So it helps with providing immunity to entrepreneurs without the fear of liability. But it's not a blanket immunity as your chapter also tries to emphasize. But it has been crucial for innovation for those online marketplaces, like your book says, Amazon, Etsy might not have existed without Section 230 because there are lawsuits and user misconduct.
So in that sense, it's been very pivotal right now. We keep talking, I know with folks in the tech industry how Section 230 might be at a tipping point because there's been a lot of bipartisan concern, if you will, over Section 230. Yeah, yeah. I think that's really important both for leadership teams to understand what it is and how it helps them now, but also to understand that there is this bipartisan effort to reduce its impact or kill it all together.
And I think as you point out, it's really the impetus for so much innovation. And I think part of the reason why the US kind of dominates in these tech platforms because they're ultimately protected. So one of the things I talk about in the book, and I'm a little, how do I say, kind of, quick with my words, when I say, when I try to explain CDA 230, to essentially protect the marketplace operators from the online and offline activity of their users, right? Obviously, Section 230 started with an online speech but has been interpreted quite broadly.
But the moment I start talking about the original legislation and how courts have interpreted it, most kind of lay people, non-lawyers, their eyes glaze over. So I just kind of truncate my description of it and kind of move on. But you had hit on, and one of the things I really want to focus on, I think people really want to learn more about, are some kind of failures and successes. In your experience, various examples, you mentioned Amazon, for example, and I would just love to just to get your take on some of these cases, what happened, what were the takeaways, and what your recommendation would be for people who are listening.
Sure. Well, I think, and you can probably provide a better sort of legal background and justification for potential liability under Section 230, but it is a very great area. So in the case of Amazon, I know firsthand that there was an incident when the eclipse in 2017 and 2024, there was a big hoopla over the solar eclipse and they were selling these sunglasses that you could kind of wear to safely watch the eclipse. However, there were issues with those glasses sold on Amazon because they were marketed as ISO compliant, but they actually lacked proper certification by unverified Chinese manufacturers.
And Amazon wasn't held liable necessarily, but the challenge there is with Section 230 is whether Amazon is deemed as a platform or a seller of those glasses. And what is that blurred line? And I know from my own discussions is that because of this incident, if you will, with these glasses, Amazon expanded its risk and compliance team to really have a more holistic approach to how they evaluate and detect potentially fraudulent sellers on the marketplace. Yeah.
It's interesting. I mentioned this in the book that Amazon is, I think, in most people's mind, a retailer, but the reality is, two-thirds of their revenue comes from their marketplace. So they're more of a marketplace than an actual retailer or straight-up e-commerce platform. And I think what's interesting about that is the kind of when these cases come down, the enormous impact that these decisions have, particularly, there have been a couple where, I think, in Texas, particularly where they have held that Amazon is liable.
And I think they're circuit splits. So I ultimately had this to probably go to the Supreme Court sooner than later. But watching Amazon and how they operate in some of these cases, I think, can be very destructive for early stage and obviously later stage, but other marketplaces because their business relies on their marketplace more or less with the volume. Yeah.
It'll be interesting in the future, too, because I know that companies trying to move into the space where TEMU and Sheen operate because they're facing growing regulatory scrutiny in TEMU and Sheen. So as they move more into that low-cost area, there's potential for more either potentially fraudulent issues that might come up with low-cost sellers. And so we'll see how that transpires. Obviously, those are goods platforms.
I think where Section 230, in my experience, has been most impactful is on services platforms. In large part, I've worked primarily in child care and in-home services where when things go wrong, they go very, very wrong. And so 230 is very helpful. I'm curious, kind of, your thoughts on, or as a platform applies 230, what things have you seen that have gone wrong or where the Section 230 may be not applies because of something the platform did to lose the protection?
So I haven't seen, based on my experience working in the tech companies, that they lost their protection. But I have seen cases where the Google Gonzalez case that went up to the Supreme Court, for example, I know when I was at the company that was concerned about what would that mean for Section 230 at the time. And I think that also touches on the broader concern about algorithms in AI prompts as we're moving into an AI-driven, centric world, including a Gentic AI, to what extent that might change the Section 230 protections. This information is another hop-on issue.
You know, Europe's DSA has a specific mis-info dis-info kind of section clause, if you will. But I know that's one area that there's a lot of bipartisan consensus over concerns about platforms and the potential bias. And so that's where I've seen it come to a head, if you will. Yeah, super interesting because obviously, AI is, I was going to say something like really updated the talk of the town, but it's everywhere, right?
And one of the things that my clients ask about is very specific uses of AI and, you know, the question is always, are there any legal concerns? Which is like the worst open-ended question you could ever ask earlier because it's like, okay, how many hours are you going to give me to research this? But my immediate response is, you know, there certainly could be some liability somewhere attached to kind of the output and, you know, if there's some harmful content that's created, et cetera. But ultimately, I think the biggest question, of course, is ownership, right?
I think the copyright office has been pretty clear that there's no ownership in the output of generative databases and tools. So more, how does the integration of an AI tool or agent impact a platform's CDA protection, right? Because with the CDA, the platform is deemed, unless they're the author, right? They have this protection and where they have not authored any of the content, they, to avail themselves of CDA 230.
What happens though when the platform integrates a, you know, an AI in a chatbot or enables users to create their profiles based on prompts or helps them communicate with one another with prompts? And so I don't know if you have any specific thoughts on that and, you know, if you've kind of noodled on that issue at all? Yeah. I think there's widespread consensus that Section 230, based on anecdotal conversations I've had, will likely change just a matter of like when, you know, and it is sort of precarious potential moment when, if it does, because of, you know, reasons you've laid out in your chapter.
But I think you touched on a really good point about algorithmic recommendations, for example, well, you know, before our chat today, I looked up a few cases where this has happened and actually, I guess TikTok, Anderson versus TikTok is a recent case where the legal debate suggests that recommendations could be considered platform-generated content if they actively promote harmful material. So you can see where this is a slippery slope. And so that was one, I guess, shot across the bow. We'll see how this evolves as you have, you know, big tech platforms, Amazon, launching and genetic AI, which could include recommendations in your shopping cart, you know, or Google, any other of these big tech companies.
There's a potential for lawsuits if, you know, AI-driven recommendations are like you said, chatbot, you know, responses could contribute to harm. And, you know, we've talked about, you know, in our community about chatbots and the youth and sort of, you know, children and sort of the types of recommendations or conversations they might have with chatbots and where that might lead. Yeah. Yeah.
It's interesting because there was a recent case, I think it's called two or three weeks ago where a user of chat GPT sued OpenAI for essentially the Famitori content that was created. OpenAI won and they said that, you know, OpenAI is not liable for the defamation. The question though is, and as you say, you know, as Section 230 changes, you know, certainly, and I guess maybe it's worth backing up a little bit, you know, Section 230 is really a procedural kind of move or procedural defense that gets the lawsuit thrown out if the plaintiffs are the defendants meet three criteria. Even if a platform integrates some sort of AI tool and they don't get the benefit of CDA to 30, it doesn't mean they lose the lawsuit.
It just means hundreds of thousands or millions of dollars and legal fees to potentially get there, right? Because obviously you still have to prove that the show causation, et cetera. And you know, the fact that a bot told, you know, gave them a recipe for arsenic instead of chicken soup is, you know, it's not great. And certainly the platform could not prevail on CDA to 30 grounds, but it doesn't necessarily mean they lose the lawsuit at the end, which I think is, I mean, the question is obviously still open, right?
But I think the interesting point and the interplay with AI is we just don't know yet. And so I hate to say to clients of mine, I don't know. But the reality is I'm watching every single AI case looking for nuggets. And even when we get the first AI case, there's going to be another circuit or district court that rules different.
And so it's going to be a while before this is figured out. But in the meantime, you know, my advice is generally just kind of go for it, you know, see what you can do, but understand that there's a risk associated with it. I don't know if you have any thoughts kind of on the future of kind of CDA to 30 protection AI or anywhere kind of related. I mean, I think like you said, we'll have to see what case law develops, right?
Similar to, I guess, my experience with auditing. Sometimes you don't know until it happens and then you develop practices from there. But I mean, you can also have catch all phrases like I noticed at Grock when I try to ask like quick assessments of certain symptoms or whatever. And they'll say, well, Grock is not a doctor.
So please consult your doctor. So I just wonder if those sort of blanket kind of statements that any kind of AI generated agent company would adopt if that's sufficient to, you know, negate potential liability. Sure. Is like a disclaimer kind of at the outset.
Yeah. Yeah. Yeah. For sure.
I think that this is, you know, if anything, it's job security for many of us because it's just such an open field right now. And certainly, you know, exciting times to be a part of it. You think you very much and that wraps up our dive into the chapter on bulletproof marketplace. If you haven't already, grab a copy of the book on Amazon.
There's a lot more detail and context in there. If you found this episode helpful, share it with your team or a fellow marketplace operator. Thanks for listening and stay bulletproof out there.