This post starts a multipart series aimed at cutting through the fog of AI hype in order to help us understand some of the real dangers we face from AI use and highlight hopeful avenues of effective resistance. It offers an overview of my argument. Following posts will take up the relevant issues in more detail.
Overview
The article critically examines the hype surrounding multimodal generative AI systems (like ChatGPT, Claude, Gemini), refuting claims of imminent Artificial General Intelligence (AGI) or a "jobs apocalypse." It argues that these systems, based on pattern recognition, are inherently limited, prone to biases, falsehoods, and "hallucinations." The shift from subscription to usage-based contracts has made them prohibitively expensive, leading businesses to curtail their use. Similarly, the revolutionary potential of AI agents, which are designed for complex tasks, is deemed oversold due to inherited flaws, unpredictable costs, and the risk of destructive decisions. The author posits that the true dangers of unchecked AI development lie in its significant costs to working people across industries, leading to job displacement and detrimental product outcomes. Furthermore, the massive infrastructure required for these systems contributes to escalating energy bills, water depletion, agricultural land loss, and exacerbates the climate crisis, while also fueling an unstable market bubble. The article distinguishes these problematic forms of AI from beneficial applications of machine learning and specialized generative AI, asserting that the fundamental issue is the class interests driving AI development to undermine human agency and economic well-being. It advocates for a unified and strengthened resistance from workers, consumers, and community groups to challenge this exploitative trajectory.
