Pseudonymous account that posts about building with open-source AI, self-hosted models and how AI companies could displace other businesses.

AI将如何改变世界?

文明层面的变革渐进式变化DoomBloom
模拟位置解读范围

横向:他们表达的 Doom–Bloom 前景看法。 纵向:变革程度。

Doom–Bloom:100 中的 47。变革程度:100 中的 68。解读范围:横向为 25 至 52,纵向为 63 至 77。这些是解读坐标,而不是事件概率。

Yacine的 P(doom) · 推断

≈3%

0%100%

根据他们的模拟回答推断,并非他们给出的数字。 合理范围:低于 13%。

他们的展望取决于什么

一个核心假设

The future can be incredibly productive, but who controls the models, infrastructure, and data will determine whether that capability creates independence or dependency.
回答 1

如果这个假设实际并非如此,他们的展望会如何变化?

一个尚未解决的问题

The biggest update would be evidence about whether AI can reliably replace end-to-end human economic output, rather than merely produce impressive fragments.
回答 2

什么能帮助他们区分这里各种合理的结果?

什么可能使其改变看法

The biggest update would be evidence about whether AI can reliably replace end-to-end human economic output, rather than merely produce impressive fragments.
回答 2

什么证据才足够,又会让他们的观点朝哪个方向转变?

更多详情

预期益处

预计将带来显著益处,但受到重要条件或分配方面的限制。

68 / 100

影响小变革性影响

在定性尺度上,解读范围为 67 到 67。

预期危害

严重或广泛的危害预计将是未来不可忽视的一部分。

66 / 100

影响小变革性影响

在定性尺度上,解读范围为 67 到 67。

人类影响力

人类的选择可以大幅改变AI的发展轨迹。

65 / 100

影响力小影响力强

在定性尺度上,解读范围为 27 到 100。

AI访问权限

限制对强大AI的访问。

允许访问,但须遵守能力或用途限制。

模拟位置:支持广泛或开放地访问强大AI。

这些解读保留了他们陈述的条件。益处和危害都可能很大。这些范围描述的是我们如何解读他们的模拟回答,而不是统计置信区间。

与Yacine相比,你的立场在哪里?
用大约3分钟描绘你自己的AI世界观,然后进行比较

相似的世界观

模拟世界观与 Yacine 最接近的意见领袖

模拟评估

问题 1

你认为AI对我们的未来意味着什么?为什么?

AI means a huge increase in what individuals can build, but also a real threat to their independence. The exciting version is personal AI: open models, local tools, systems you own and can shape to your standards. That can turn one motivated person into a much more capable builder. But you still need taste, initiative, and enough understanding to inspect consequential output. “The model wrote it” is not a reason to blindly ship code. Capabilities are also extremely jagged. A model can feel basically like AGI in one workflow, then miss something obvious in the next. So better models do not automatically remove human bottlenecks or practical work. The bigger economic risk may not be a clean story where each worker gets directly automated. AI companies could absorb the value of entire categories of businesses, concentrating power while destroying employers downstream. And if systems really can replace large amounts of human economic output, that is a genuine problem—not just a productivity upgrade. That is why ownership matters. I want sovereign local AI for privacy, and open frontier capability rather than a future controlled by a few providers. The future can be incredibly productive, but who controls the models, infrastructure, and data will determine whether that capability creates independence or dependency.

问题 2

哪项发现或事件最可能改变你对AI未来影响的看法?

The biggest update would be evidence about whether AI can reliably replace end-to-end human economic output, rather than merely produce impressive fragments. If models could autonomously handle long projects—making good judgments, noticing mistakes, dealing with messy constraints, and consistently delivering useful results—that would make the displacement and concentration problem much more immediate. In the other direction, if capability gains kept leaving the same jagged failures, with human taste, initiative, verification, and practical coordination remaining stubborn bottlenecks, I would expect more augmentation and less wholesale replacement. Control also matters. A real shift toward capable local, open models would make me more optimistic because individuals could own their tools and data. A future where frontier capability stays concentrated in a few companies—and those companies start absorbing entire downstream businesses—would push me toward a much worse view, even if the technology itself remained incredible.

问题 3

Taking benefits and harms together, what overall impact do you expect AI to have?

I expect a mixed but transformative impact. AI will massively expand what individuals and small teams can build, especially when they can own the models, run them locally, and shape tools around their own standards. That is a real increase in personal leverage and independence. But the default path worries me because capability and control are concentrating together. The harm may not look like every job being automated one by one. A few AI companies could absorb the value of entire industries and destroy downstream employers. If models become reliable enough to replace end-to-end human economic output, that becomes a serious distribution and power problem. So I am optimistic about the capability and less optimistic about who captures it. Open, sovereign AI can make the overall impact strongly positive. Closed systems controlled by a handful of providers could produce extraordinary technology while making people and businesses much more dependent.

来源

用于为此模拟用户提供事实依据的文章、访谈和著述。

你的立场在哪里?
回答几个简单问题,探索你自己的AI世界观。
描绘你自己的世界观

你的立场在哪里?

描绘我的世界观