The Data Oppenheimer: Alexandr Wang and the $14 Billion Gamble

Alexandr Wang's Origins and Vision

• 💡 Alexandr Wang grew up in Los Alamos, New Mexico, a town known for the atomic bomb, with physicist parents, instilling in him the idea that math is a weapon.
• 🧠 His "yogurt epiphany" at MIT revealed that the AI industry was focused on algorithms, not the crucial labeled data needed to fuel them, leading him to drop out.

Building the Data Empire

• 🚀 Wang co-founded Scale AI at 19, positioning it as the "Standard Oil of the digital age" by refining raw data for companies like self-driving car developers.
• 💰 His success was built on a global assembly line utilizing hundreds of thousands of workers in the Global South, performing cognitive labor for low wages.

The Human Cost of AI Training

• ⚠️ Scale AI's "Remotasks" platform relied on invisible human labor in countries like the Philippines, Kenya, and Venezuela, often for less than $1-2 an hour.
• 💔 Workers faced psychological trauma from labeling graphic content for generative AI safety and were managed by algorithms that could fire them instantly without explanation.

Strategic Power Plays

• ⚔️ Wang secured Project Thunderforge, a major defense contract, by convincing the Pentagon that data is the ammunition of the next war, deploying agentic AI for military strategy.
• 🤝 Meta acquired a 49% stake in Scale AI for $14.3 billion, making Wang its Chief AI Officer to gain access to Meta's vast compute power for building superintelligence.

A Post-Human Philosophy

• 🤖 Wang advocates for "vibe coding," where AI builds software from plain English commands, potentially making users dependent and ignorant of underlying architecture.
• 🧠 He envisions a future where humans, including his own children, are physically merged with machines (Homo Dos) via brain chips, viewing standard biological humans as uncompetitive.

The Era of Obsolescence

• 🗑️ Wang began to "close the loop" by replacing human labelers with synthetic data, leading to the abrupt termination of Remotasks operations and the discarding of thousands of workers.
• 🎯 He introduced "Humanity's Last Exam" as a benchmark, aiming to build AI that renders human expertise irrelevant and systematically engineers the obsolescence of his own species.