For the past three years, artificial intelligence has largely been seen as a tool for cutting costs—automating tasks, streamlining operations, and boosting efficiency. But a growing number of founders and investors believe we are now entering AI’s “Second Wave,” a new era defined not by optimization, but by the creation of entirely new products and experiences that simply couldn’t exist before large language models. Kylan Gibbs, former Google DeepMind product manager and CEO of AI startup Inworld, describes the shift clearly: “The first wave of AI made existing things cheaper. Automation. Efficiency. The next wave makes things that couldn’t exist before—new products, new experiences, new revenue.” For Gibbs, this distinction is critical. If AI only reduces expenses, it redistributes value within existing industries. But if it enables novel consumer offerings—products people are willing to pay for—it expands the economic pie. To support this vision, Gibbs launched a Silicon Valley accelerator in January, backing up to 30 startups focused on building consumer-scale AI experiences rather than just improving legacy workflows. The initiative includes support from major venture firms like Khosla Ventures and Lightspeed Venture Partners, as well as leaders from OpenAI, Google, and Stripe. A demo day is set for early March in San Francisco. This mindset aligns with a recent message from Y Combinator CEO Garry Tan: “Instead of worrying about doing the same thing we’ve been doing for cheaper, why not focus on doing the thing we never even dreamed of doing?” Several startups are already embodying this new era. Particle, an AI-native news platform founded by Sara Beykpour, exemplifies the shift. What once took a month to develop can now be built, tested, and launched in hours. The team even jokes about “boomer thinking” when someone defaults to old ways of working. One of their key innovations is Podcast Clips—AI-curated snippets from long-form audio, automatically linked to relevant news stories using embeddings. Instead of searching through hours of content, users get the most pertinent moments, with AI-generated summaries and context. This transforms how information is consumed, making it more dynamic and personalized. Meanwhile, Luvu, a fitness app co-founded by Alexis Sursock and Creston Brooks, uses generative AI to create a highly personalized training experience. Its AI “marshmallow” sends real-time, context-aware messages—like “Your test is over. Time to work out!”—that adapt to a user’s life. These messages have a click-through rate four times higher than generic prompts. Retention is also dramatically improved, with two to three times more users staying active after 30 days compared to industry averages. The app uses computer vision to analyze workouts and provide instant feedback, while also leveraging reinforcement learning with verified rewards to refine its AI. This creates a feedback loop between user behavior and AI improvement—something only possible with modern models. Then there’s Status, an AI-powered social simulation game founded by Fai Nur. With over 3 million downloads, the app lets users role-play as characters in dynamic, AI-generated social worlds—whether as Hogwarts students, soccer stars, or characters from Stranger Things. Posts trigger real-time, AI-driven responses from other characters, and actions have consequences. An “aura score” tracks performance, and players level up or face backlash based on their choices. In traditional settings, the unpredictable nature of LLMs is a drawback. But in gaming, that variability is a strength—creating fresh, engaging, and evolving experiences. For pioneers like Gibbs, the Second Wave isn’t about making things faster or cheaper. It’s about unlocking new forms of interaction, creativity, and connection. The future of AI won’t just be in boardrooms and back offices—it will be in apps people love, games they play, and experiences they pay for.