AI's Future: Between Annihilation and Salvation?

The future of artificial intelligence is often framed in extreme terms: either a path to utopia or an apocalyptic end, potentially involving a scenario where AI turns humanity into paperclips. This thought experiment, conceived by philosopher Nick Bostrom, illustrates the potential for an AI with a singular goal—like maximizing paperclip production—to disregard human existence and repurpose atoms for its objective.

Eleanor Drage, in her book "What If We Got AI Right?", contends that these polarized views obscure a more nuanced reality. AI has become a pervasive buzzword, utilized by startups for funding, by companies for product marketing, and by filmmakers for dystopian narratives. Drage highlights the significant influence of major technology companies like Google, Meta, and OpenAI, suggesting they are shaping public perception of AI's future to align with their own interests.

Drage, an AI ethics researcher, posits that the unfettered pursuit of profit is the primary driver of AI's dangers, rather than hypothetical rogue AI. She argues that AI is more likely to exacerbate existing problems, such as the climate crisis. For instance, Google's data center in Dalles, Oregon, consumes over 350 million gallons of water annually, contributing significantly to the city's water usage. Projections indicate that by 2027, the AI sector could consume up to 134 terawatt-hours of energy per year, a figure comparable to the Netherlands' total energy consumption.

As a solution, Drage advocates for a move away from dependence on Big Tech. She argues that the development of AI requires broader input beyond just technologists, emphasizing the need for individuals with common sense and life experience. Drage questions the current model, asking, "What if AI wasn't built by Big Tech but by local communities?" She asserts that shaping AI's development should not be solely left to Big Tech companies and supposed experts, as all individuals have a stake in its future.