AI’s Unpopularity + Competing With ChatGPT — With Olivia Moore

Episode Summary

Executive Summary: Olivia Moore argues that AI’s public backlash reflects fear, media narratives, and lab leaders’ dystopian messaging, but adoption will keep rising as consumers and businesses see real utility. She says big chatbot companies won’t win everything: constrained labs leave room for vertical, opinionated startups, especially where integrations, workflows, and memory matter. OpenClaw-like agents and memory-first products will reshape work, but mainstream value will come from specialized AI tools, not a single super-app.

Main Topics: Public fear vs. practical AI value (Priority: 5/5): The conversation opens on why AI sentiment is so negative despite widespread use. Moore attributes this to media scare stories, job-loss rhetoric from lab leaders, and uncertainty about AI’s trajectory, while noting consumer usefulness will gradually outweigh fear. Will AI be winner-take-all? (Priority: 5/5): They debate whether ChatGPT, Claude, Gemini, and other large chatbots will capture most value. Moore argues the market will be broad, not winner-take-all, because frontier labs are resource-constrained and focused on their own priorities. Where startups can still compete (Priority: 5/5): Moore explains that startups can win in vertical, opinionated products, especially where broad chatbots are not optimized for specific workflows, accuracy requirements, or legacy integrations. Agentic software and OpenClaw (Priority: 4/5): A major theme is the rise of autonomous agents that can take actions across apps and workflows. Moore sees this as a major architecture shift, but says consumer use is still limited and developers are the strongest users. Memory, personalization, and super-app behavior (Priority: 4/5): They discuss persistent memory as a key differentiator in consumer AI, enabling products to know users deeply and potentially serve as login, context, and orchestration layers across other apps. Creative tools, image/video, and platform collapse (Priority: 4/5): The discussion covers how image generators were quickly absorbed by bigger models, while video and social AI remain less settled. Moore notes that some categories are getting commoditized faster than others. Implications for incumbents and the SaaS landscape (Priority: 4/5): The episode closes on how incumbents are adapting, whether AI-native startups will displace legacy software, and how a future of AI-first companies could reshape business creation and competition.

Key Arguments: AI sentiment is negative largely because the public hears repeated warnings about water use, job displacement, and existential risk, not because people lack direct utility from the tools. Adoption will improve as consumers experience benefits firsthand; ChatGPT already has massive scale, and mainstream usage should soften skepticism over time. Lab leaders may sound alarmist, but the current reality is that humans still direct most AI use; autonomy is increasing, but AI is not yet replacing full human judgment across most tasks. Startups can still build durable businesses because frontier labs are constrained by compute, inference, and focus, and many opportunities sit outside the core priorities of ChatGPT, Claude, and Gemini. Vertical AI products can beat general-purpose chatbots when they need specific interfaces, regulated outputs, accuracy guarantees, or deeply embedded integrations with legacy systems. Memory is becoming a key moat and product layer: AI that remembers preferences, context, and prior interactions can deliver dramatically better experiences and become a new software front door. OpenClaw-style agents are most compelling for developers and operators who want long-running, asynchronous task automation; mainstream consumer adoption is still limited. The biggest AI-native value may come not from a chatbot super-app, but from many specialized companies built on top of the same foundation models. AI often increases, rather than reduces, work: leverage allows users to ship more, start more projects, and intensify output rather than simply cut hours. Incumbents are waking up and building AI features, but AI-native startups still have an opening because new founders will often choose the ground-up AI version over legacy software.

Data Points: U.S. voter sentiment toward AI: 57% - NBC News poll cited in the opening; voters said risks outweigh benefits. Net sentiment toward AI: -20 - Overall positive vs. negative sentiment in the cited poll. ChatGPT users: 900 million - Moore cited ChatGPT’s scale while discussing why negativity persists despite adoption. Usage gap between average users and power users: 8-9x - She said AI power users utilize the tools far more than average users. ChatGPT vs. Gemini web usage gap: 2.5-3x - Moore described ChatGPT still leading Gemini by a substantial margin on web usage. ChatGPT vs. Claude web usage gap: ~30x - She said ChatGPT’s usage is still far ahead of Claude. App overlap between ChatGPT and Claude app ecosystems: 11% overlap - Used to illustrate divergent product strategies and ecosystems. OpenClaw followers achieved in experiment: 1,000 - Moore described an agent-driven X account she created that reached 1,000 followers before getting banned. Initial followers before viral boost: 100 - The account grew organically to 100 before being amplified by the crypto community. Sora downloads: 1 million - Moore noted Sora reached 1 million downloads faster than ChatGPT. Sora daily active users: 3 million - She said Sora still has roughly 3 million daily active users. Google standalone products on report list: 4 - Google had four standalone AI products in her report: Gemini, NotebookLM, AI Studio, and Google Labs. OpenClaw-style web rank in February: #10 to #30 (traffic down/flat) - She said OpenClaw would have ranked around 10 in one snapshot but its traffic flattened, suggesting limited consumer expansion.

Pivotal Quotes: "every tech company is going to be an AI company, and every AI company is going to be an agent company" — Olivia Moore: Her thesis on the long-term reinvention of the technology industry. "I think that every tech company is going to be an AI company, and every AI company is going to be an agent company." — Olivia Moore: Repeated framing of AI as a platform shift rather than a single-product market. "the models are amazing. I will talk to founders who are like, 11's expensive, I'm gonna switch to this instead. And then they always switch back because the quality of the voices is just so much better." — Olivia Moore: On ElevenLabs as an example of a startup defensible through quality and head start.

Implications: AI winners will likely be a mix of giant labs and many vertical startups. For users, memory-rich, agentic, AI-first tools will reshape work and software onboarding; for companies, adopting AI quickly may become a survival advantage.