달의 이성 학원 · Sample · Student
Modern AI companies often present their systems as neutral tools—pipelines of math and engineering that simply “scale.” Yet every stage of that pipeline encodes choices: which data is scraped, which labor is paid to label it, which markets are prioritized, and which risks are treated as acceptable externalities.
Critics argue that calling this process “innovation” can hide accountability. When a model amplifies bias or concentrates power, the public is told that technology is inevitable. Readers of Empire of AI are asked to follow incentives, not only algorithms—and to notice who benefits when systems grow faster than the institutions meant to govern them.
Multiple choice
Multiple choice
Short answer