Original post · AI translation
点击这里查看。
一次精彩的发布,为人工智能走完最后一公里做好准备。
每家实验室都能抓取同一个互联网的数据,而互联网上几乎没有记录资深专家是如何做决策的。
新近推出的 @arcfront 将企业几十年来保存的专家决策过程记录,转化为前沿人工智能模型的训练数据。
这些记录来自 300 多家企业合作伙伴,这比从网上抓取的任何东西都更难复制。
称其在融资不到 200 万美元后,不到一年就实现了 10 亿美元的年化收入
Arcfront 称,5 家前沿人工智能实验室中已有 4 家付费购买其关于医生、工程师和银行从业者如何作出艰难决策的逐步记录。
Original
A brilliant launch to prepare AI for the last mile. Every lab can scrape the same internet, and the internet barely records how a senior expert decides. Newly launched @arcfront takes something companies have kept for decades, records of how their experts reach decisions, and turns it into training data for frontier AI models. Those records come through 300+ enterprise partnerships, which are much harder to copy than anything scraped from the web. says it hit a $1B annual revenue run rate in under a year after raising less than $2M Arcfront says 4 of the 5 frontier AI labs already pay it for step-by-step records of how doctors, engineers and bankers make hard calls.
今天,我们与 @Naman_Bhalla 一起,在 @asxna 的大力支持下推出 @arcfront。
不到一年,5 家前沿实验室中有 4 家成为了客户。
2025 年,一款人工智能在 ICPC 世界总决赛上解出了全部 12 道题。成绩最好的人类队伍解出了 11 道。我参加过其中两届总决赛,所以我知道这是多么了不起的成就。
然而,让同一款人工智能在公司里担任财务经理,或者从事几乎任何其他工作,它在第 1 天就会失败。
这是为什么?
Original
Today we're launching @arcfront with @Naman_Bhalla and deep support from @asxna. In under a year, four of the five frontier labs became customers. In 2025 an AI solved all 12 problems at the ICPC World Finals. The best human team solved 11. I competed in two of those finals, so I know how big a feat that is. And yet, put that same AI inside a company as a finance manager, or in almost any other job, and it fails on day 1. Why is that?