Fireside Chat, Interview, Other
2024 Big Ideas: AI-First Games That Never End with Jonathan Lai
- Core Thesis: A16Z partner Jonathan Lai predicts 2024 will launch the first cohort of "AI-first games that never end," leveraging Large Language Models (LLMs) to reinvent game mechanics rather than just optimizing development efficiency.
- Strategic Mantra: The guiding philosophy for this shift is "good alone, great with AI, and best with friends," positioning AI companions as core gameplay elements rather than mere training bots.
- Two-Wave AI Revolution in Gaming:
- Wave 1 (Current Reality): Existing studios use AI tools (ChatGPT, Midjourney, Adobe Firefly) to accelerate production; 87% of 230 surveyed studios are already utilizing these tools, rating their impact as "transformational."
- Wave 2 (Future Opportunity): Development of entirely new "AI-native" games featuring dynamic worlds, emergent NPC behavior, and real-time neural network rendering.
- Technological Capabilities:
- Generative agents will serve as lifelike, personalized companions that learn from user preferences and adapt to game contexts (e.g., combat or stealth).
- Narrative systems will enable unique, non-repeating playthroughs where the game world is generated at runtime based on player actions.
- Game onboarding is poised for reinvention through AI-driven, personalized tutorials.
- Market Confidence Factors:
- Consumer demand for AI companions is validated by historical precedents like Cortana in Halo and Ellie in The Last of Us, which have fostered deep player attachment despite being pre-scripted.
- Hyper-personalization is an established market trend, evidenced by avatar customization in Baldur's Gate 3, skin economies in Fortnite/League of Legends, and the improvisational nature of Dungeons & Dragons.
- Incumbent vs. Startup Dynamics:
- Large AAA publishers (e.g., Activision, Microsoft) are likely to benefit from Wave 1 efficiency gains but may struggle to innovate on Wave 2 due to their reliance on existing IP and sequels.
- New entrants and startups are projected to lead the "AI-native" shift, similar to how mobile gaming disrupted the console/PC market dominated by companies like EA and Activision.
- Emerging AI experiences (e.g., Character.ai) may blur the line between games and other consumer engagement, attracting demographics outside traditional gaming.
- Business Model Evolution:
- Traditional models (Premium $60 upfront or Free-to-Play with microtransactions) face profitability challenges due to the high inference costs of serving dynamic AI worlds.
- Subscription models are identified as the likely default, where users pay a recurring fee to cover the variable costs of AI processing, mirroring successes in services like Xbox Game Pass and Character.ai.
- Developers must design mechanics to balance "time versus money," potentially using virtual currencies or energy meters to manage inference costs, a precedent established by games like FarmVille.
- Call to Action for Builders: Lai encourages developers to reimagine existing genres (MMO, sports, chess, poker) from first principles by integrating generative AI into core gameplay and to share prototypes, noting that major hits like Dota and Counter-Strike originated as small-team mods.
- Market Scope: Successfully executing this shift offers the potential to disrupt a $200 billion market, which A16Z notes is the fastest-growing category in media.