问题 1
Demis Hassabis
x.com/demishassabisGoogle DeepMind chair who works on AI for scientific discovery and medicine and favors independent safety standards and international coordination.
AI将如何改变世界?
横向:他表达的 Doom–Bloom 前景看法。 纵向:变革程度。
Doom–Bloom:100 中的 74。变革程度:100 中的 81。解读范围:横向为 69 至 79,纵向为 75 至 100。这些是解读坐标,而不是事件概率。
一个核心假设
So my positive expectation is conditional on substantial work: scientifically grounded evaluations that evolve with capabilities, independent standards, international coordination and a willingness to slow frontier development if evidence warrants it.回答 2
如果这个假设实际并非如此,他的展望会如何变化?
一个尚未解决的问题
I do not attach a precise probability to catastrophic outcomes, but I regard the risk as nonzero and non-negligible.回答 1
什么能帮助他区分这里各种合理的结果?
更多详情
仍有几种解读是合理的:预计将带来具有变革性且广泛有价值的收益。 / 预计将带来显著益处,但受到重要条件或分配方面的限制。
85 / 100
在定性尺度上,解读范围为 67 到 100。
严重或广泛的危害预计将是未来不可忽视的一部分。
66 / 100
在定性尺度上,解读范围为 67 到 67。
人类的选择可以大幅改变AI的发展轨迹。
70 / 100
在定性尺度上,解读范围为 50 到 75。
预计AI仍将是能力有限的工具。
预计AI将在大多数认知工作中达到人类水平。
模拟位置:预计AI将在认知工作中大幅超越人类。
停止或大幅放缓开发能力更强的AI。
模拟位置:在落实所述保障措施的前提下继续开发。
加快开发能力更强的AI。
这些解读保留了他陈述的条件。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的模拟回答,而不是统计置信区间。
相似的世界观
模拟世界观与 Demis Hassabis 最接近的意见领袖
Demis Hassabis关于AI说过的话
Hassabis expects AGI to transform science and medicine and calls for cautious progress, frontier safety standards and international coordination.
“Dario’s essay points towards the right path forward. The details need working through, but the direction is correct for meeting this critical moment.”
Post on X “When there is a large degree of uncertainty and the stakes are this high, proceeding with cautious optimism is the sensible and correct strategy.”
Essay, A Framework for Frontier AI and the Dawning of a New Age “Urgent action is needed to address risks that might arise as we get closer to AGI.”
Essay, A Framework for Frontier AI and the Dawning of a New Age “AGI has the potential to be the ultimate tool for advancing science and medicine, and to drive enormous productivity gains and economic growth.”
Essay, A Framework for Frontier AI and the Dawning of a New Age “I’ve always believed the No.1 application of AI should be to improve human health.”
Post on X
逐字引自所链接的出处,核对于 2026年10月3日
来源
用于为此模拟用户提供事实依据的文章、访谈和著述。
Organizer transcript, distributed as a February podcast of the January Davos discussion. Hassabis distinguishes verifiable coding and mathematics from experiment-limited science and forming new hypotheses. Wants time and scientific cooperation to solve technical safety and societal adaptation. Attribute only his turns, not Amodei’s faster forecasts.

Older mechanism and values foundation. Hassabis rejects a precise P(doom) while describing risk as nonzero and non-negligible, calls for much more scientific safety work, and distinguishes misuse from autonomous loss of control. Prefers collaborative research to a weapons race and sees medicine, energy and abundance as transformative opportunities.

Originally published in July; the institute republishes the essay. Forecasts transformative AGI within a few years, proposes evolving independent technical tests and a standards body, and allows coordinated slowdown. The institute distinguishes author views from official Google policy.

模拟评估
问题 2
综合考虑益处和危害,你预计AI会对我们的社会产生怎样的总体影响?
问题 3
你认为AI最终会在多大程度上改变世界?
问题 4
凭直觉,你认为AI导致人类灭绝或类似的永久性灾难的概率有多大?
你的立场在哪里?
描绘我的世界观


