Interview, Fireside Chat, Podcast
The State of Consumer Tech in the Age of AI
Market Dynamics & Velocity
- Consumer AI is in an early era where "velocity" (both distribution and product iteration) is the primary competitive moat, replacing the network effects that defined previous decades.
- Unlike the mature platforms of the mobile/cloud era (where features were incremental), AI requires "relentless model updates" to maintain market position; companies risk becoming obsolete if they fall behind in capability rather than just feature sets.
- Consumer spam is predicted to evolve into "food, rent, software" consumption models, where software subsumes discretionary spend on entertainment, creativity, and relationship intermediation.
Business Models & Monetization
- Top-tier AI consumer products command significantly higher prices ($200–$250/month) compared to the historical $50/year average for subscription apps, as they replace high-value labor (e.g., Deep Research replacing 10 hours of manual work).
- Revenue retention rates in AI are meaningfully higher than unique user retention because users upgrade usage tiers via credits, overages, and premium features, whereas traditional subscriptions had static pricing.
- Enterprise adoption is outpacing mass consumer adoption for several products (e.g., Eleven Labs); enterprise buyers often discover tools via consumer viral trends on social platforms and adopt them for internal strategy.
The State of Social & Connection
- A dedicated "AI social graph" has not yet emerged; current AI creative content remains distributed on legacy platforms (Reddit, Instagram, TikTok) rather than native AI networks.
- Defensibility in AI social networks is challenged by "skeuomorphic" designs that mimic legacy feeds (e.g., AI-generated photo feeds) rather than creating new interaction modalities.
- Potential new social paradigms include "synthetic selves" (AI clones of real people) that allow users to scale their wisdom or personas, and AI companions that serve as therapy, coaching, or conversation partners.
- Companion apps are identified as a critical, enduring use case that may actually facilitate better real-world human connection by providing low-stakes practice for social interaction, rather than replacing it.
Form Factors & Hardware
- Mobile remains the primary substrate, but there is a strong push toward "edge" AI (local LLMs) for privacy and always-on capabilities.
- AirPods and similar wearable audio devices are identified as the most likely immediate hardware expansion for AI, though social norms around "always-on" recording must evolve.
- Future "agentic" models are expected to move beyond suggestions to taking direct action (e.g., sending emails, booking services) via devices that can "see" and "hear" user contexts.
- AR glasses and other head-mounted displays are considered long-term possibilities, but current adoption is lagging behind software innovation; local model development is a prerequisite for device-level AI.
Talent & Content Creation
- The creator economy is fragmenting into two distinct paths: "human-centric" celebrities (e.g., Taylor Swift, where personal life narrative matters) and "interest-based" AI agents (e.g., AI influencers focused on niche topics).
- AI art and video are not devalued by ease of generation; high-quality output still requires significant time and curation, maintaining a low conversion rate for top-tier talent.
- Audio/voice technology is becoming a critical enterprise interface, replacing offshore call centers with synthetic voices that handle complex negotiations and sales pitches with higher compliance and consistency.
Future Outlook
- While some early AI social attempts may fail (potentially becoming "MySpace" equivalents), companies that maintain a state-of-the-art model frontier and ship rapidly will persist, even in a competitive landscape.
- The "peak value" of AI is framed as enabling better human connection, potentially reducing isolation and anxiety for users who lack traditional support systems.
- New social norms regarding recording and data privacy are emerging organically, similar to the cultural adjustments seen when mobile phones were introduced, particularly among younger generations who treat digital recording as standard practice.