Predicting machine learning moats and the automation of work are two important topics that intersect when considering the future of technology and business. Understanding how machine learning systems develop enduring moats and how automation affects the job market can provide valuable insights into the direction industries are heading.

Kazuki

Hatched by Kazuki

Aug 18, 2023

3 min read

0

Predicting machine learning moats and the automation of work are two important topics that intersect when considering the future of technology and business. Understanding how machine learning systems develop enduring moats and how automation affects the job market can provide valuable insights into the direction industries are heading.

When it comes to machine learning moats, the key lies in the interface between scaling laws and products. While software scales with zero marginal costs, machine learning scales with nonlinear emergent behaviors. This means that in order to build a truly great business, one must focus on creating structural advantages beyond just the model itself.

Data, in particular, plays a crucial role in creating moats for machine learning systems. Well-defined and curated training data that is diverse and not repeated when scaling can provide lasting advantages. Unlike models, which can easily be replaced with procedural changes, data is not easily taken by leaving employees or leaked. Companies like Runway and Jasper understand the importance of data and are crafting moats in their respective verticals.

However, it's important to note that moats alone are not enough. Benedict Evans reminds us that every wave of automation brings about the disappearance of certain jobs but also creates new ones. The misconception that there is a fixed amount of work to be done, and that automation will result in less work for people, is known as the Lump of Labour fallacy. History has shown that as automation progresses, we move up the scale of human capability and new job categories emerge.

Innovation and automation can actually lead to more jobs, thanks to the Jevons Paradox. This paradox suggests that as we make tools more efficient and cheaper to run, we end up using more of them for new and different things. This, in turn, creates a demand for more labor. Automation doesn't necessarily replace jobs; it often leads to the creation of new job opportunities.

However, integrating transformative technologies into large, complex companies can be a challenge. Startups operate on an 18-month funding cycle, while enterprises follow an 18-month decision cycle. This misalignment can hinder the adoption and implementation of innovative technologies in established organizations. Bridging this gap requires understanding the needs and constraints of both startups and enterprises.

In conclusion, predicting machine learning moats and understanding the impact of automation on the job market are crucial exercises for businesses and industries. By focusing on the interface between scaling laws and products, companies can build enduring moats beyond just the model itself. Data plays a vital role in creating these moats, providing lasting advantages when well-defined and diverse. Additionally, it's important to recognize that automation doesn't necessarily lead to job loss. Instead, it creates new job opportunities as we move up the scale of human capability. To fully leverage the benefits of automation, bridging the gap between startups and enterprises is essential. This can be achieved by aligning funding cycles with decision cycles and understanding the needs of both sides.

Hatch New Ideas with Glasp AI 🐣

Glasp AI allows you to hatch new ideas based on your curated content. Let's curate and create with Glasp AI :)