Dec 29, 2025 · 1h 0m · mixergy

#2290 He’s building an AI media empire

Ricardo Vice Santos · 43m spoken Andrew Warner · 11m spoken
0:00 / 0:00

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

In this interview with Andrew Warner, Dream Stories founder Ricardo Vice Santos explains how he builds AI-powered personalized children's books and shares his vision for a future where digital media is custom-generated for individual consumption.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Andrew holds 21.4% of the talking time here. How this is scored →

Andrew as informed peer 4.5 Guest teaching 4.8 Guest disagreement 2.2 Andrew pushing back 3.3
05100:0015:0030:0045:001:00:000:32–5:58 · Andrew as informed peer 5/10 Scaling Book Creation and Addressing Revenue Metrics Andrew presses Ricardo immediately on revenue figures and tries to frame the company as a get-rich-quick success story, citing past metrics. Ricardo firmly rejects that framing, refusing to talk about revenue to keep the focus on long-term vision. Andrew also tests Ricardo's Spotify backstory by citing Matt Farley's handmade novelty songs.5:58–8:00 · Andrew as informed peer 2/10 Disappointment with Children's Media and Tintin Inspiration Ricardo explains his motivation for founding Dream Stories, contrasting his childhood love for Tintin with modern children's media that he views as mundane lowest-common-denominator content. Andrew listens supportively with minimal pushback.8:00–12:05 · Andrew as informed peer 4/10 Solving Image Consistency in AI Book Generation Andrew demonstrates familiarity with generative image tools like ChatGPT and Nano Banana, describing the character drift problem across multiple scenes. Ricardo expands with technical details on using LoRAs, Stable Diffusion, and filtering poor-quality customer inputs.12:06–14:50 · Andrew as informed peer 2/10 Partnering with Phil and the Future of Media Generation Ricardo recounts teaming up with his co-founder Phil and outlines his thesis that zero marginal creation cost will unlock self-consumed bespoke media rather than just cheaper Hollywood productions. Andrew clarifies the core distinction of creating content purely for personal consumption.14:51–22:02 · Andrew as informed peer 4/10 Buyer Psychology and the Bedtime Ritual Ricardo uses the Gillette razor model to explain products where the purchaser is not the end consumer, justifying why he targeted parents buying bedtime books rather than direct consumer video networks. Andrew synthesizes the concept with his own Suno song experiments.22:02–27:59 · Andrew as informed peer 4/10 Sponsor Spotlight: PR Automation Using Zapier Following an ad read and PR automation discussion via Zapier, Ricardo explains the compute unit economics making multi-character books viable while video remains prohibitively expensive due to high inference costs. Andrew validates the long-term vision of values-aligned storytelling.28:00–33:05 · Andrew as informed peer 6/10 Distribution Challenges and Ad Unit Economics Andrew directly quotes Ricardo's past interview with Nathan Latka regarding CAC, AOV, and Facebook ad buying, pressing for deeper operational insight. Ricardo reframes Andrew's points, explaining the dangers of targeting 1.0 ROAS on physical products without guaranteed repeat purchases.33:05–41:50 · Andrew as informed peer 6/10 Rethinking Subscriptions and Developing a Push Service Andrew highlights a paradox in Ricardo's advice: criticizing subscribe-and-save models while Dream Stories itself charges per transaction without an active subscription tier. Ricardo defends his stance by explaining that subscription is earned only when the product shifts from pull creation to push-based developmental milestones.41:51–47:41 · Andrew as informed peer 7/10 AI Personalization Opportunities in Enterprise and Coaching Andrew takes the lead by proposing a detailed B2B product concept: automated executive coaching distilled from meeting transcripts and leadership books via Zapier. Ricardo validates the idea, linking it to his own workflow with NotebookLM and discussing how context management rather than commodity LLMs creates defensibility.47:41–54:42 · Andrew as informed peer 4/10 Customer Support Insights and Synthetic Data Training Andrew challenges Ricardo to name a concrete insight from doing customer support daily, skeptically noting the business seems straightforward. Ricardo educates him on discovering synthetic data training: users uploaded previous AI outputs to avoid blurry photos, which enabled Dream Stories to reduce input requirements to a single image.54:43–58:35 · Andrew as informed peer 5/10 Implementing Intercom's Fin and AI Agent Personas Andrew discusses Intercom's positioning transition into an AI agent platform with Fin and requests Ricardo's internal persona guidelines. Ricardo explains why customer support AI must embody an engineered brand persona rather than mimicking the founder's voice.0:32–5:58 · Guest teaching 4/10 Scaling Book Creation and Addressing Revenue Metrics Andrew presses Ricardo immediately on revenue figures and tries to frame the company as a get-rich-quick success story, citing past metrics. Ricardo firmly rejects that framing, refusing to talk about revenue to keep the focus on long-term vision. Andrew also tests Ricardo's Spotify backstory by citing Matt Farley's handmade novelty songs.5:58–8:00 · Guest teaching 3/10 Disappointment with Children's Media and Tintin Inspiration Ricardo explains his motivation for founding Dream Stories, contrasting his childhood love for Tintin with modern children's media that he views as mundane lowest-common-denominator content. Andrew listens supportively with minimal pushback.8:00–12:05 · Guest teaching 5/10 Solving Image Consistency in AI Book Generation Andrew demonstrates familiarity with generative image tools like ChatGPT and Nano Banana, describing the character drift problem across multiple scenes. Ricardo expands with technical details on using LoRAs, Stable Diffusion, and filtering poor-quality customer inputs.12:06–14:50 · Guest teaching 4/10 Partnering with Phil and the Future of Media Generation Ricardo recounts teaming up with his co-founder Phil and outlines his thesis that zero marginal creation cost will unlock self-consumed bespoke media rather than just cheaper Hollywood productions. Andrew clarifies the core distinction of creating content purely for personal consumption.14:51–22:02 · Guest teaching 5/10 Buyer Psychology and the Bedtime Ritual Ricardo uses the Gillette razor model to explain products where the purchaser is not the end consumer, justifying why he targeted parents buying bedtime books rather than direct consumer video networks. Andrew synthesizes the concept with his own Suno song experiments.22:02–27:59 · Guest teaching 4/10 Sponsor Spotlight: PR Automation Using Zapier Following an ad read and PR automation discussion via Zapier, Ricardo explains the compute unit economics making multi-character books viable while video remains prohibitively expensive due to high inference costs. Andrew validates the long-term vision of values-aligned storytelling.28:00–33:05 · Guest teaching 6/10 Distribution Challenges and Ad Unit Economics Andrew directly quotes Ricardo's past interview with Nathan Latka regarding CAC, AOV, and Facebook ad buying, pressing for deeper operational insight. Ricardo reframes Andrew's points, explaining the dangers of targeting 1.0 ROAS on physical products without guaranteed repeat purchases.33:05–41:50 · Guest teaching 6/10 Rethinking Subscriptions and Developing a Push Service Andrew highlights a paradox in Ricardo's advice: criticizing subscribe-and-save models while Dream Stories itself charges per transaction without an active subscription tier. Ricardo defends his stance by explaining that subscription is earned only when the product shifts from pull creation to push-based developmental milestones.41:51–47:41 · Guest teaching 5/10 AI Personalization Opportunities in Enterprise and Coaching Andrew takes the lead by proposing a detailed B2B product concept: automated executive coaching distilled from meeting transcripts and leadership books via Zapier. Ricardo validates the idea, linking it to his own workflow with NotebookLM and discussing how context management rather than commodity LLMs creates defensibility.47:41–54:42 · Guest teaching 7/10 Customer Support Insights and Synthetic Data Training Andrew challenges Ricardo to name a concrete insight from doing customer support daily, skeptically noting the business seems straightforward. Ricardo educates him on discovering synthetic data training: users uploaded previous AI outputs to avoid blurry photos, which enabled Dream Stories to reduce input requirements to a single image.54:43–58:35 · Guest teaching 4/10 Implementing Intercom's Fin and AI Agent Personas Andrew discusses Intercom's positioning transition into an AI agent platform with Fin and requests Ricardo's internal persona guidelines. Ricardo explains why customer support AI must embody an engineered brand persona rather than mimicking the founder's voice.0:32–5:58 · Guest disagreement 5/10 Scaling Book Creation and Addressing Revenue Metrics Andrew presses Ricardo immediately on revenue figures and tries to frame the company as a get-rich-quick success story, citing past metrics. Ricardo firmly rejects that framing, refusing to talk about revenue to keep the focus on long-term vision. Andrew also tests Ricardo's Spotify backstory by citing Matt Farley's handmade novelty songs.5:58–8:00 · Guest disagreement 2/10 Disappointment with Children's Media and Tintin Inspiration Ricardo explains his motivation for founding Dream Stories, contrasting his childhood love for Tintin with modern children's media that he views as mundane lowest-common-denominator content. Andrew listens supportively with minimal pushback.8:00–12:05 · Guest disagreement 1/10 Solving Image Consistency in AI Book Generation Andrew demonstrates familiarity with generative image tools like ChatGPT and Nano Banana, describing the character drift problem across multiple scenes. Ricardo expands with technical details on using LoRAs, Stable Diffusion, and filtering poor-quality customer inputs.12:06–14:50 · Guest disagreement 2/10 Partnering with Phil and the Future of Media Generation Ricardo recounts teaming up with his co-founder Phil and outlines his thesis that zero marginal creation cost will unlock self-consumed bespoke media rather than just cheaper Hollywood productions. Andrew clarifies the core distinction of creating content purely for personal consumption.14:51–22:02 · Guest disagreement 2/10 Buyer Psychology and the Bedtime Ritual Ricardo uses the Gillette razor model to explain products where the purchaser is not the end consumer, justifying why he targeted parents buying bedtime books rather than direct consumer video networks. Andrew synthesizes the concept with his own Suno song experiments.22:02–27:59 · Guest disagreement 1/10 Sponsor Spotlight: PR Automation Using Zapier Following an ad read and PR automation discussion via Zapier, Ricardo explains the compute unit economics making multi-character books viable while video remains prohibitively expensive due to high inference costs. Andrew validates the long-term vision of values-aligned storytelling.28:00–33:05 · Guest disagreement 3/10 Distribution Challenges and Ad Unit Economics Andrew directly quotes Ricardo's past interview with Nathan Latka regarding CAC, AOV, and Facebook ad buying, pressing for deeper operational insight. Ricardo reframes Andrew's points, explaining the dangers of targeting 1.0 ROAS on physical products without guaranteed repeat purchases.33:05–41:50 · Guest disagreement 3/10 Rethinking Subscriptions and Developing a Push Service Andrew highlights a paradox in Ricardo's advice: criticizing subscribe-and-save models while Dream Stories itself charges per transaction without an active subscription tier. Ricardo defends his stance by explaining that subscription is earned only when the product shifts from pull creation to push-based developmental milestones.41:51–47:41 · Guest disagreement 1/10 AI Personalization Opportunities in Enterprise and Coaching Andrew takes the lead by proposing a detailed B2B product concept: automated executive coaching distilled from meeting transcripts and leadership books via Zapier. Ricardo validates the idea, linking it to his own workflow with NotebookLM and discussing how context management rather than commodity LLMs creates defensibility.47:41–54:42 · Guest disagreement 3/10 Customer Support Insights and Synthetic Data Training Andrew challenges Ricardo to name a concrete insight from doing customer support daily, skeptically noting the business seems straightforward. Ricardo educates him on discovering synthetic data training: users uploaded previous AI outputs to avoid blurry photos, which enabled Dream Stories to reduce input requirements to a single image.54:43–58:35 · Guest disagreement 1/10 Implementing Intercom's Fin and AI Agent Personas Andrew discusses Intercom's positioning transition into an AI agent platform with Fin and requests Ricardo's internal persona guidelines. Ricardo explains why customer support AI must embody an engineered brand persona rather than mimicking the founder's voice.0:32–5:58 · Andrew pushing back 6/10 Scaling Book Creation and Addressing Revenue Metrics Andrew presses Ricardo immediately on revenue figures and tries to frame the company as a get-rich-quick success story, citing past metrics. Ricardo firmly rejects that framing, refusing to talk about revenue to keep the focus on long-term vision. Andrew also tests Ricardo's Spotify backstory by citing Matt Farley's handmade novelty songs.5:58–8:00 · Andrew pushing back 1/10 Disappointment with Children's Media and Tintin Inspiration Ricardo explains his motivation for founding Dream Stories, contrasting his childhood love for Tintin with modern children's media that he views as mundane lowest-common-denominator content. Andrew listens supportively with minimal pushback.8:00–12:05 · Andrew pushing back 2/10 Solving Image Consistency in AI Book Generation Andrew demonstrates familiarity with generative image tools like ChatGPT and Nano Banana, describing the character drift problem across multiple scenes. Ricardo expands with technical details on using LoRAs, Stable Diffusion, and filtering poor-quality customer inputs.12:06–14:50 · Andrew pushing back 2/10 Partnering with Phil and the Future of Media Generation Ricardo recounts teaming up with his co-founder Phil and outlines his thesis that zero marginal creation cost will unlock self-consumed bespoke media rather than just cheaper Hollywood productions. Andrew clarifies the core distinction of creating content purely for personal consumption.14:51–22:02 · Andrew pushing back 3/10 Buyer Psychology and the Bedtime Ritual Ricardo uses the Gillette razor model to explain products where the purchaser is not the end consumer, justifying why he targeted parents buying bedtime books rather than direct consumer video networks. Andrew synthesizes the concept with his own Suno song experiments.22:02–27:59 · Andrew pushing back 1/10 Sponsor Spotlight: PR Automation Using Zapier Following an ad read and PR automation discussion via Zapier, Ricardo explains the compute unit economics making multi-character books viable while video remains prohibitively expensive due to high inference costs. Andrew validates the long-term vision of values-aligned storytelling.28:00–33:05 · Andrew pushing back 5/10 Distribution Challenges and Ad Unit Economics Andrew directly quotes Ricardo's past interview with Nathan Latka regarding CAC, AOV, and Facebook ad buying, pressing for deeper operational insight. Ricardo reframes Andrew's points, explaining the dangers of targeting 1.0 ROAS on physical products without guaranteed repeat purchases.33:05–41:50 · Andrew pushing back 6/10 Rethinking Subscriptions and Developing a Push Service Andrew highlights a paradox in Ricardo's advice: criticizing subscribe-and-save models while Dream Stories itself charges per transaction without an active subscription tier. Ricardo defends his stance by explaining that subscription is earned only when the product shifts from pull creation to push-based developmental milestones.41:51–47:41 · Andrew pushing back 2/10 AI Personalization Opportunities in Enterprise and Coaching Andrew takes the lead by proposing a detailed B2B product concept: automated executive coaching distilled from meeting transcripts and leadership books via Zapier. Ricardo validates the idea, linking it to his own workflow with NotebookLM and discussing how context management rather than commodity LLMs creates defensibility.47:41–54:42 · Andrew pushing back 5/10 Customer Support Insights and Synthetic Data Training Andrew challenges Ricardo to name a concrete insight from doing customer support daily, skeptically noting the business seems straightforward. Ricardo educates him on discovering synthetic data training: users uploaded previous AI outputs to avoid blurry photos, which enabled Dream Stories to reduce input requirements to a single image.54:43–58:35 · Andrew pushing back 3/10 Implementing Intercom's Fin and AI Agent Personas Andrew discusses Intercom's positioning transition into an AI agent platform with Fin and requests Ricardo's internal persona guidelines. Ricardo explains why customer support AI must embody an engineered brand persona rather than mimicking the founder's voice.

speaking balance: gold is Andrew, purple is the guest (3 minute bins)

0:00 · Andrew 41.1% · guest 58.9%0:00 · Andrew 41.1% · guest 58.9%3:00 · Andrew 18.2% · guest 81.8%3:00 · Andrew 18.2% · guest 81.8%6:00 · Andrew 24.9% · guest 75.1%6:00 · Andrew 24.9% · guest 75.1%9:00 · Andrew 10.2% · guest 89.8%9:00 · Andrew 10.2% · guest 89.8%12:00 · Andrew 13.9% · guest 86.1%12:00 · Andrew 13.9% · guest 86.1%15:00 · Andrew 38% · guest 62%15:00 · Andrew 38% · guest 62%18:00 · Andrew 10.3% · guest 89.7%18:00 · Andrew 10.3% · guest 89.7%21:00 · Andrew 26.5% · guest 73.5%21:00 · Andrew 26.5% · guest 73.5%24:00 · Andrew 19.1% · guest 80.9%24:00 · Andrew 19.1% · guest 80.9%27:00 · Andrew 13.7% · guest 86.3%27:00 · Andrew 13.7% · guest 86.3%30:00 · Andrew 15.4% · guest 84.6%30:00 · Andrew 15.4% · guest 84.6%33:00 · Andrew 24.9% · guest 75.1%33:00 · Andrew 24.9% · guest 75.1%36:00 · Andrew 11.9% · guest 88.1%36:00 · Andrew 11.9% · guest 88.1%39:00 · Andrew 15.5% · guest 84.5%39:00 · Andrew 15.5% · guest 84.5%42:00 · Andrew 54.9% · guest 45.1%42:00 · Andrew 54.9% · guest 45.1%45:00 · Andrew 18.1% · guest 81.9%45:00 · Andrew 18.1% · guest 81.9%48:00 · Andrew 5.5% · guest 94.5%48:00 · Andrew 5.5% · guest 94.5%51:00 · Andrew 20% · guest 80%51:00 · Andrew 20% · guest 80%54:00 · Andrew 17.4% · guest 82.6%54:00 · Andrew 17.4% · guest 82.6%57:00 · Andrew 22.4% · guest 77.6%57:00 · Andrew 22.4% · guest 77.6%1:00:00 · Andrew 100% · guest 0%1:00:00 · Andrew 100% · guest 0%
Sharpest disagreement ▶ 1:21 Guest rejects revenue discussion and get-rich-quick angle

Ricardo immediately shuts down Andrew's revenue question and rejects the premise of turning his startup into a get-rich-quick story, demanding they focus on the product vision.

Hardest push from Andrew ▶ 35:44 Host challenges guest on subscription pricing contradiction

Andrew directly catches the tension between Ricardo's strong criticism of subscribe-and-save funnels and the fact that Dream Stories relies on one-off transactional sales.

Biggest teaching moment ▶ 50:30 Guest explains synthetic data training derived from customer hacks

After Andrew expresses skepticism that Ricardo learns anything new from support tickets, Ricardo delivers an in-depth breakdown of how customer photo workarounds revealed a synthetic data pipeline that cut input requirements to one photo.

Andrew holds their own ▶ 43:27 Host pitches complete B2B personalized coaching system

Andrew demonstrates deep operational knowledge by architecting an automated B2B executive coaching pipeline using meeting transcripts, Zapier, and core company management texts.

the scores for every segment, with the reasoning behind each
ChapterTopicAndrew as informed peerGuest teachingGuest disagreementAndrew pushing backWhy
Scaling Book Creation and Addressing Revenue Metrics 5456 Andrew presses Ricardo immediately on revenue figures and tries to frame the company as a get-rich-quick success story, citing past metrics. Ricardo firmly rejects that framing, refusing to talk about revenue to keep the focus on long-term vision. Andrew also tests Ricardo's Spotify backstory by citing Matt Farley's handmade novelty songs.
Disappointment with Children's Media and Tintin Inspiration 2321 Ricardo explains his motivation for founding Dream Stories, contrasting his childhood love for Tintin with modern children's media that he views as mundane lowest-common-denominator content. Andrew listens supportively with minimal pushback.
Solving Image Consistency in AI Book Generation 4512 Andrew demonstrates familiarity with generative image tools like ChatGPT and Nano Banana, describing the character drift problem across multiple scenes. Ricardo expands with technical details on using LoRAs, Stable Diffusion, and filtering poor-quality customer inputs.
Partnering with Phil and the Future of Media Generation 2422 Ricardo recounts teaming up with his co-founder Phil and outlines his thesis that zero marginal creation cost will unlock self-consumed bespoke media rather than just cheaper Hollywood productions. Andrew clarifies the core distinction of creating content purely for personal consumption.
Buyer Psychology and the Bedtime Ritual 4523 Ricardo uses the Gillette razor model to explain products where the purchaser is not the end consumer, justifying why he targeted parents buying bedtime books rather than direct consumer video networks. Andrew synthesizes the concept with his own Suno song experiments.
Sponsor Spotlight: PR Automation Using Zapier 4411 Following an ad read and PR automation discussion via Zapier, Ricardo explains the compute unit economics making multi-character books viable while video remains prohibitively expensive due to high inference costs. Andrew validates the long-term vision of values-aligned storytelling.
Distribution Challenges and Ad Unit Economics 6635 Andrew directly quotes Ricardo's past interview with Nathan Latka regarding CAC, AOV, and Facebook ad buying, pressing for deeper operational insight. Ricardo reframes Andrew's points, explaining the dangers of targeting 1.0 ROAS on physical products without guaranteed repeat purchases.
Rethinking Subscriptions and Developing a Push Service 6636 Andrew highlights a paradox in Ricardo's advice: criticizing subscribe-and-save models while Dream Stories itself charges per transaction without an active subscription tier. Ricardo defends his stance by explaining that subscription is earned only when the product shifts from pull creation to push-based developmental milestones.
AI Personalization Opportunities in Enterprise and Coaching 7512 Andrew takes the lead by proposing a detailed B2B product concept: automated executive coaching distilled from meeting transcripts and leadership books via Zapier. Ricardo validates the idea, linking it to his own workflow with NotebookLM and discussing how context management rather than commodity LLMs creates defensibility.
Customer Support Insights and Synthetic Data Training 4735 Andrew challenges Ricardo to name a concrete insight from doing customer support daily, skeptically noting the business seems straightforward. Ricardo educates him on discovering synthetic data training: users uploaded previous AI outputs to avoid blurry photos, which enabled Dream Stories to reduce input requirements to a single image.
Implementing Intercom's Fin and AI Agent Personas 5413 Andrew discusses Intercom's positioning transition into an AI agent platform with Fin and requests Ricardo's internal persona guidelines. Ricardo explains why customer support AI must embody an engineered brand persona rather than mimicking the founder's voice.

Statements from this episode (18)

Assertion Not checkable as stated
Dream Stories has generated about 70,000 user characters since launch
“But we've had since we launched, we had about 70,000 characters created on the platform.”
Ricardo Vice Santos Dec 29, 2025 ▶ 0:54
Insight
Santos: Giving people creation tools creates novel behaviors, not linear extrapolations
“And so I think if you give people the tools to creation, they will do things that are not necessarily an extrapolation of what we do today.”
Ricardo Vice Santos Dec 29, 2025 ▶ 5:25
Opinion
Santos: Modern children's media caters to the lowest common denominator
“To be honest, a bit more mundane and there, there's content, I think it has high production value, but it feels a sort of like, for lack of a better expression, lowest common denominator to, you know, it's kind of like something that's very agreeable, you know…”
Ricardo Vice Santos Dec 29, 2025 ▶ 6:57
Insight
Santos: Building consumer AI requires consistency without multiple tries
“It's an entirely different thing to build a product for other people to use it, like, you know, the, you know, with sort of some sort of symbols of quality and consistency, right? So again, you actually mentioned this, right? You go in, you do it, and eventual…”
Ricardo Vice Santos Dec 29, 2025 ▶ 9:34
Disclosure
Santos: Dream Stories used ChatGPT, Stable Diffusion, and LoRAs for consistency
“We use a number of things. We, at the time we used the chat to PT. We used the stable diffusion. I forget the first model we used. We used to do it. We used LORAS, which was we still do in some regards but LORAS is something that, you know, that you literally …”
Ricardo Vice Santos Dec 29, 2025 ▶ 10:22
Prediction Not checkable as stated
Santos: Zero-cost AI will create media made entirely for self-consumption
“I'm more of a believer that there's going to be sort of new behaviors that you can't see just yet. And this is one of them, right? Which is if cost near zero, then literally everyone can do it. But if everyone can do it, then of course, you know, the, who's go…”
Ricardo Vice Santos Dec 29, 2025 ▶ 14:30
Assertion Contradicted
Santos: Women make roughly 70% of Gillette razor purchases
“There's a few products that are like this, like the one I usually tell is like, is Gillette where, you know, usually the woman in the relationship is the one that buys the Gillette. It's like, I think about 70% of the purchases of Gillette are by women. Don't,…”
Ricardo Vice Santos Dec 29, 2025 ▶ 16:50
Disclosure
Vice Santos: Dream Stories sells multiple books per customer via episodic sequences
“I told you that we sell multiple books, you know, to the same person you know, essentially because they're episodic, right? It's a sequence.”
Ricardo Vice Santos Dec 29, 2025 ▶ 21:11
Assertion Not checkable as stated
Santos: OpenAI's Sora is reportedly burning $15 million per day
“This is why it's kind of known that, you know, Sora, you know, OpenAI Sora is literally burning, it's been said to be burning fifteen million dollars per day, right, on the app, essentially with people like playing with the video.”
Ricardo Vice Santos Dec 29, 2025 ▶ 24:53
Opinion
Santos: AI video generation is currently too expensive to be viable
“Video is actually, It's still expensive enough that it's hard to actually come up with a, if you're not a hyperscaler, to come up with a viable business model. Arguably, it's not a business model even for them.”
Ricardo Vice Santos Dec 29, 2025 ▶ 25:08
Insight
Vice Santos: Multi-character consistency in generative AI is exponentially harder
“If you look at most comparable products out there usually they have one character just because the whole thing about having, if it's already hard to keep consistent with one character character, it becomes actually exponentially harder with multiple, not becau…”
Ricardo Vice Santos Dec 29, 2025 ▶ 26:32
Insight
Santos: AI tools make product distribution harder and more critical than execution
“Particularly in the time where you have all these tools like you know, vibe coding, lovable, so on, so forth, where everyone can create the product. You can, you know, it, it's a great time in the sense that, you know, if you ever, you know, if you were ever t…”
Ricardo Vice Santos Dec 29, 2025 ▶ 28:30
Disclosure
Santos: Dream Stories will not make ABC books due to zero repeatability
“That's why we're not doing ABC books. You know, people ask me to say, well, can I get a book about ABC? I was like, no, not interested in that, right? There's no repeatability to that.”
Ricardo Vice Santos Dec 29, 2025 ▶ 32:54
Insight
Santos: Subscriptions must be a feature of a service, not a discount hook
“What you want to do is you want to build like a service, something that actually is a service in which Subscription is just a feature of that service.”
Ricardo Vice Santos Dec 29, 2025 ▶ 34:25
Assertion Not checkable as stated
Santos: A subscription-only checkout converted better than offering one-off purchases
“Do an A-B test, you know, have a funnel with subscription only and the funnel where you give both choices and then see what the conversion does. And lo and behold, the one with subscription actually increased conversion because it reduced that because again, y…”
Ricardo Vice Santos Dec 29, 2025 ▶ 37:30
Assertion Supported
Warner: Zapier's founder automates employee meeting feedback based on management books
“Wade, the founder, likes the, what is it, the five dysfunctions of an organization, of a team. I forget the name of the book. Anyway, he likes it so much that that, at the end of each internal meeting, Everyone's individual contribution to the meeting gets ana…”
Andrew Warner Dec 29, 2025 ▶ 43:20
Insight
Santos: Dream Stories trains consistent models using synthetic photo variations
“We were able to reduce the number of pictures required to literally just one, because then we're like, holy , you know, we can essentially just ask for one picture, generate the synthetic version of it, and then say, do you like this? Great. So we generate a f…”
Ricardo Vice Santos Dec 29, 2025 ▶ 54:13
Opinion
Santos: Intercom's Fin works well only with extensive tailoring and customization
“Now it's not amazing. But you know, but it's good and but it requires a lot of customization, right? And I would say that I would still recommend it. But I think the value, you know, if you go out and try Finn, I think what you're going to be very disappointed…”
Ricardo Vice Santos Dec 29, 2025 ▶ 55:49
Made with StarZero

Turn any episode into a week of clips.

This entire site, about 40 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.