OpenAI: Strength and Confidence
Sure! Here’s a revised version of the content with external promotional call-to-actions removed or rephrased:
With that said, Empire of AI is a powerful work, brimming not only with insightful reporting but also with significant ideas. This shines through in two main themes.
The first is straightforward: it tells the story of ambition overriding ethics. The history of OpenAI, as described by Hao (and echoed by Hagey), reflects a company founded on the idealistic pursuit of creating a safety-focused artificial general intelligence, which ultimately became more interested in competition. This narrative is familiar in the realm of Big Tech. Consider past cases like Theranos, aiming to simplify diagnostics, or Uber, which sought to disrupt the traditional taxi industry. However, the closest parallel might be Google, transitioning from its "Don’t be evil" mantra to being viewed as an illegal monopolist. Furthermore, examine how Google shifted from exercising caution in releasing its language model to hastily launching a chatbot to compete with OpenAI. In Silicon Valley, regardless of initial intentions, the focus often returns to winning.
The second theme is more intricate and forms the book’s thesis regarding what Hao terms AI colonialism. This concept posits that major AI companies operate like traditional empires, extracting wealth from the lower tiers of society through labor, creative contributions, and resources, thereby enriching those at the top. “I’ve found only one metaphor that encapsulates the nature of what these AI power players are: empires,” she states. “During the long era of European colonialism, empires seized and extracted resources not their own, exploiting the labor of those they subjugated to mine, cultivate, and refine those resources for their enrichment.” She also chronicles her growing disillusionment with the industry: “With increasing clarity,” she notes, “I realized that the very revolution promising to bring a better future was instead, for people on the margins of society, reviving the darkest remnants of the past.”
To illustrate this, Hao steps away from her desk to observe the effects of this empire as it expands globally. She travels to Colombia to meet data labelers tasked with teaching AI about various images, one of whom she describes rushing back to her apartment for the chance to earn a few dollars. She documents how workers in Kenya handling content moderation for OpenAI were left traumatized by the disturbing material they encountered. In Chile, she highlights how the industry extracts valuable resources—such as water, power, copper, and lithium—to build data centers.
