...
【24h】

A plea for intelligent nanotechnology journalism

机译:对智能纳米技术新闻的呼吁

获取原文
获取原文并翻译 | 示例
           

摘要

Ray Kurzweil's evocative "singularity" is that epoch when artificial intelligence first overtakeshuman intelligence.~1 It is intimately associated with nanotechnology as an enabler ofexponentially growing information processing power (as exemplified by Moore's law). Basedon present trends of technology development, albeit difficult to assess when those trends areexponential or even superexponential, Kurzweil estimates that the singularity will occur around2050. This presumably assumes that human intelligence will remain at roughly the same level.Some trends might lead one to surmise that it is actually in decline. If so, would the singularityoccur even sooner, since it should be easier for machines to surpass humans? On the other handit might become progressively more difficult to sustain the required technological developmentof the machines (which remain necessarily dependent on humans until the singularity occurs).The two effects might compensate each other. An early singularity, however, implies that theabsolute level of computational capability achieved might then be too low to enable the post-singularity expansion of intelligence throughout the universe to be realized.
机译:雷·库兹韦尔(Ray Kurzweil)令人回味的“奇异性”是人工智能首次超过人类智能时的那个时代。〜1它与纳米技术密切相关,成为信息处理能力呈指数增长的推动力(例如摩尔定律)。根据目前的技术发展趋势,尽管很难评估这些趋势是指数级的还是什至超指数级的,库兹韦尔估计,奇异性将在2050年左右发生。这大概是假设人类智力将保持在大致相同的水平上。某些趋势可能导致人们猜测它实际上正在下降。如果是这样,那么奇异性是否会更快出现,因为机器应该更容易超越人类?另一方面,要维持所需的机器技术发展可能会变得越来越困难(直到出现奇异之处,机器仍必须依赖于人类)。这两种影响可能会相互抵消。然而,早期的奇异性意味着所达到的计算能力的绝对水平可能太低,以至于无法实现整个宇宙的智能的奇异性后扩展。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号