Dear Commons Community,
I have just finished reading Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI by Karen Hao. I found it an interesting and at times a troubling read learning about the personalities of those leading AI development at OpenAI. Hao paints Altman as a duplicitous character who will say or do anything to make OpenAI the number one AI company in the world. She gives numerous examples of his lying to colleagues. Chapters 4 (Dreams of Modernity), 12 (Plundered Earth), 14 (Deliverance), and 16 (Cloak and Dagger) were particularly illuminating and in the case of the latter three, disturbing. At 400 plus pages, it is not a short book and can get very technical at times. I had to keep my iPhone close to me whenever I was reading it to look up unfamiliar terms.
As a nod to poetic justice, below is a review produced by the AI program Co-Pilot, a derivative of OpenAI’s ChatGPT.
Tony
———————————————————-
Empire of AI is a gripping exposé that blends investigative journalism with global storytelling to unpack the rise of OpenAI and the broader implications of artificial intelligence.
📚 Overview
- The book chronicles OpenAI’s transformation from a nonprofit idealist to a corporate powerhouse backed by Microsoft, now valued at over $300 billion.
- Hao draws on 300 interviews, including 90 with insiders, to reveal the internal culture, ethical tensions, and environmental costs of AI development.
- She opens with the dramatic 2023 boardroom coup that temporarily ousted CEO Sam Altman, using it as a lens to explore power dynamics and governance failures.
🌍 Key Themes
- Digital Colonialism: Hao likens the AI industry to a modern empire, extracting labor and resources from the Global South while concentrating profits in Silicon Valley.
- Environmental Toll: Training large models demands massive energy and water, with data centers consuming resources equivalent to small cities.
- Hidden Labor: Workers in Kenya, Venezuela, and the Philippines perform low-paid, often traumatic tasks like content moderation and data cleaning.
- Transparency vs. Secrecy: OpenAI’s shift from openness to opacity is a central critique, with Hao arguing that basic facts about AI systems are now obscured.
🧠 Analytical Depth
- Hao doesn’t just critique OpenAI—she questions the entire AI development culture that prioritizes speed and dominance over ethics and inclusivity.
- She explores how AI models reflect the biases of their creators and often fail those outside the Silicon Valley bubble.
- The book also touches on the philosophical fervor within OpenAI, where employees oscillate between utopian dreams and existential dread about AGI.
✍️ Writing Style
- Hao’s prose is clear, engaging, and emotionally resonant. She balances technical clarity with human stories, making complex topics accessible without oversimplifying.
- Her tone is critical but grounded, offering a nuanced view of the dilemmas facing AI developers.
🧭 Verdict
Empire of AI is essential reading for anyone interested in the intersection of technology, ethics, and global power. It’s not just a story about OpenAI—it’s a cautionary tale about unchecked ambition, hidden costs, and the urgent need for democratic oversight in AI development.
This entry is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license.
