World's Fair 2025
On Curiosity — Sharif Shameem, Lexica
Overview
This talk argues that curiosity is the primary driving force for innovation, enabling the translation of future ideas into present realities. The speaker emphasizes that building and sharing demos is the most effective way to explore the potential of AI models, as these models are not fully understood even by their creators. Demos serve as a crucial tool for AI engineers, akin to excavators uncovering hidden capabilities, with curiosity acting as the guide.
Who should watch
- AI engineers exploring new model capabilities.
- Product managers seeking to understand the potential of AI for new products.
- Builders experimenting with generative models and AI tools.
- Anyone interested in the process of AI discovery and innovation.
- Individuals feeling stuck or unsure about their next steps in AI development.
Key takeaways
- Curiosity, often felt as an intuition or subconscious pattern recognition, is the main force pulling ideas from the future into the present.
- Demos are the most effective way to explore and understand the capabilities of AI models, which are complex and not fully predictable.
- Building demos allows engineers to uncover latent capabilities within AI models, even with existing technology, potentially leading to significant product development over the next decade.
- AI engineering is more akin to excavation than traditional goal-oriented engineering, focusing on discovering hidden potential within models.
- Early AI models had severe limitations, such as small context windows and restrictive terms of service, yet demos were still built to explore their possibilities.
- The speaker highlights historical examples, like Henri Poincaré's breakthrough in fusion functions and Charles Darwin's extensive study of barnacles, to illustrate how deep, seemingly unrelated exploration can lead to significant discoveries.
- There is a moral obligation to explore and share findings through demos, honoring the pioneers of computing and advancing the field.
- Uncertainty and exploration are core to AI engineering; if you know exactly what you're doing, you might be missing opportunities for discovery.
Notable quotes
*The role of this unconscious work in mathematical invention appears to me as incontestable.* - paraphrased from Henri Poincaré
*In science, if you know what you're doing, you should not be doing it. Engineering if you know what you're doing you should not be doing it.* - Richard Hamming
*Your unique perspective shouldn't be wasted, and I think you have a moral responsibility to share them with the world.*
Unofficial community note. Prefer the recording for nuance.