Latest / The Series A Show with Fexingo: Early-Stage Funding, Pitch Decks, and Startup Milestones / Why Series A Pitches Now Include an AI Trust Slide
Transcript
- Lucas: So, Luna, I've been looking at a bunch of Series A decks over the past few weeks, and there's a new slide showing up that I don't think we saw even six months ago. It's the 'AI trust' slide. Luna: AI trust? Like a slide saying 'trust us, we're responsible'? Lucas: That's exactly what I would have expected. But it's actually more specific. It's a slide that demonstrates how the startup handles ai generated content, data provenance, and model governance. And it's becoming a real point of differentiation in the current funding environment. Luna: What's driving that? Is it just the Anthropic watermark news, or is it broader? Lucas: Well, Anthropic's been making waves with their watermarking and content provenance features for Claude. They're giving creators a way to certify that images or text were ai generated. But the bigger context is that AI backlash is becoming a 'crisis of trust,' as Anthropic's CEO put it. Investors are starting to ask: what happens when your product gets accused of generating misinformation or deepfakes? Luna: And that's a direct threat to the business model, not just a PR issue. Lucas: Right. And it's not hypothetical. There was that TechCrunch story just yesterday about a woman whose stepfather used Grok to transform a childhood photo into explicit imagery. That kind of thing makes regulators and customers nervous. So founders are being asked: how do you prevent your tool from being misused? What's your accountability framework? Luna: So the AI trust slide is basically a risk-mitigation slide. But how does a founder actually structure it? Lucas: The good ones I've seen aren't just a list of principles. They're showing concrete mechanisms. Like, 'we watermark every output using a standard that can be verified by third parties.' Or 'we have a content moderation pipeline that flags potentially harmful generations before they're delivered.' It's about showing, not telling. Luna: And that resonates with investors because it shows the founder has thought about the product's lifecycle, not just the demo. Lucas: Exactly. And it's becoming part of due diligence. We're seeing some funds now have a checklist item for AI governance, similar to data compliance. If you don't have a clear answer, it's a red flag. Luna: And by the way, this is exactly the kind of conversation that makes this show useful. If you're building or running a company, these are the questions your investors are going to ask. And listener support is what keeps these conversations free. If you find value in what we're doing here, you can support the show at buy me a coffee dot com slash fexingo. Lucas: Yeah, it's a small gesture that helps us keep doing this. So, back to the trust slide. Another layer is around data provenance. Founders are showing where their training data comes from, how they handle copyrighted material, and whether they have licenses in place. Luna: That feels like a minefield. I mean, there's still so much legal uncertainty around training data. Lucas: Exactly, and that's why investors are so interested. It's a potential liability that could be existential. So a strong trust slide will include a data audit trail — like 'we source training data from licensed partners, and we can prove it.' Luna: That makes sense. But isn't there a risk that this becomes just another box-ticking exercise? Lucas: There is. I've seen a couple of decks where the trust slide feels bolted on — generic language, no specifics. But the best ones are tied directly to the product's unique risks and mitigation. And they often include a slide on model governance, like who reviews outputs, how you handle bias, and what your escalation process is. Luna: And I'd imagine that's especially important for startups in sensitive domains like health or finance. Lucas: Absolutely. And there's a market signal here. Look at NVIDIA — up 3.5% over the last five days, while the broader tech complex is down. That tells you where the money is still flowing. But with that comes responsibility. And Series A investors are increasingly asking: are you building AI responsibly enough to survive a scandal? Luna: So how does a founder actually craft that slide? What are the key elements? Lucas: First, you need a clear statement of your AI usage. Not just 'we use AI' — but where, and for what purpose. Second, you need to show your safeguards: watermarking, content filters, human-in-the-loop reviews. Third, you need to describe your governance structure, like who owns AI ethics at the company. And finally, you should tie it to industry standards or certifications you're pursuing. Luna: That sounds like a lot for one slide. Lucas: It is, but the best ones are almost like an infographic — not text-heavy. They might use icons and short phrases to convey the key points, and then the founder can elaborate in the Q&A. Because the slide is really just a conversation starter. Luna: And I imagine the trust slide is becoming a way to signal maturity. Like, we're not just a bunch of engineers hacking away in a garage. Lucas: Exactly. And it also aligns with the data compliance slide we talked about in episode 153, but it goes further. Compliance is about meeting legal requirements; trust is about meeting ethical and stakeholder expectations. They're complementary. Luna: So if I'm a founder, where do I start? I can't just copy Anthropic's approach. Lucas: Right, you have to tailor it to your use case. But the good news is there are emerging frameworks and tools. For example, the Coalition for Content Provenance and Authenticity — C2PA — has standards for digital content provenance. Many AI companies are adopting that. So you can say: 'we generate C2PA-compliant metadata for every ai generated asset.' That's concrete and verifiable. Luna: That's helpful. And I bet investors also want to see how you handle user data, not just training data. Lucas: Yes, that's part of it. The trust slide should also cover data privacy — how you collect, use, and protect user data. And whether you have robust security measures in place. Because if you're handling sensitive data, that's a major trust component. Luna: So the trust slide is really a holistic view of how you handle data and models. Lucas: It is. And it's becoming a differentiator. In a competitive funding round, a strong trust slide can set you apart. It shows you've anticipated questions and have answers ready. It's a signal of founder quality. Luna: And what about the actual pitch? How much time should a founder spend on the trust slide? Lucas: I'd say maybe 30 seconds in the main deck, but you have to be prepared to go deep in Q&A. It's not the hook; it's the foundation. So you mention it confidently, then move on. But if an investor asks a follow-up, you need to have all the details at your fingertips. Luna: That makes sense. And are there any examples of startups that have done this really well? Lucas: Well, without naming specific companies that might not be around in a year, I can tell you that the ones that impress me are the ones that already have a public-facing trust page on their website. They treat trust as a product feature, not a slide. And they often have one of those 'responsible AI' tabs that explains their principles and practices. Luna: So it's really about embedding trust into the company's DNA. Lucas: Exactly. And that's what investors are looking for. They want to know that you're not just paying lip service, but that you've actually built the infrastructure to maintain trust at scale. Because the moment you launch, you're in the public eye. Luna: And with the AI backlash we're seeing, that's only going to become more important. Lucas: Right. So my advice to founders: don't treat the trust slide as an afterthought. Invest time in it. Make it specific, make it credible, and make it a true reflection of how you operate. Because in a few years, it'll be as standard as the team slide. Luna: And maybe it'll even help you win over skeptical investors who've seen too many AI pitches go wrong. Lucas: That's the hope. And the more we see these slides become the norm, the better the whole ecosystem will be. Luna: Thanks, Lucas. That was a super practical rundown. Lucas: Always a pleasure. And to our listeners, if you're working on a Series A deck, now's the time to start thinking about your trust slide. It could be the difference between a yes and a no. Luna: If you found this helpful, you know where to find us. Until next time. Lucas: Stay sharp, and keep building responsibly.