人工智能:均衡器还是分裂者?

1作者: borissk18 天前
俗话说:“上帝创造了人,但塞缪尔·柯尔特使他们平等。”人工智能有时也能产生类似的效果。在我见过的一个企业系统中,引入人工智能代理后,那些每周使用某工具 40 小时的人与每周只使用 2 小时的人之间的差距几乎消失了。我还见过,在每个人都获得了慷慨的 Cursor 订阅后,顶尖开发者(10 倍效率)与普通开发者之间的差距也缩小了。 与此同时,近期关于 Anthropic 的 Claude Fable 5 和 Mythos 模型公开报道表明,最强大的系统可能通过有限访问计划或高级渠道推出,而不是广泛而平均地提供。 纵观历史,谁能负担得起当时最先进技术的费用,往往决定了谁能掌权。在古希腊,重装步兵是自备装备的普通自由公民,将兵役与公民参与联系起来。同样,殖民时期的美洲民兵也将武装服务融入了日常社区生活。相比之下,中世纪晚期欧洲昂贵的盔甲、战马和军事训练费用,将军事力量牢牢掌握在少数富人手中——这导致了雅典和早期美国的民主传统,却加剧了中世纪欧洲的专制统治。 因此,在我看来,人工智能是成为一种“拉平者”,还是成为一种加剧拥有者和非拥有者之间差距的工具,这是一个重要的问题。
查看原文
There is a saying “God made men, but Samuel Colt made them equal.” AI sometimes can have similar effect. In one enterprise system I’ve seen, the gap between people using a tool 40 hours a week and those using it 2 hours a week nearly disappeared after AI agents were introduced. I have also seen the gap between a 10x developer and an average one shrink after everyone got generous Cursor subscriptions.<p>At the same time, recent public reporting on Anthropic’s Claude Fable 5 and Mythos-class models suggests that the most capable systems may be rolled out through limited-access programs or premium channels rather than broadly and evenly.<p>Throughout history, who could afford access to the best technology of the time often shaped who got to govern. In ancient Greece, hoplites were ordinary free citizens who funded their own gear, tying military service to civic participation. Similarly, colonial American militias embedded armed service into everyday community life. By contrast, the prohibitive cost of armor, warhorses and military training in late medieval Europe placed military power firmly in the hands of the wealthy few - leading to democratic traditions in Athens and the early United States, while reinforcing autocratic rule across medieval Europe.<p>So IMHO it is an important question whether AI becomes an equalizer, or a tool that widens the gap between those who have access and those who do not.