888 X 動態摘要｜06/05

生成時間: 2026-06-05 06:26:34

總結

AI產業在技術創新、人才需求與安全監管方面均有顯著動態，預示著關鍵的市場機遇與風險。

今日重點

1. AI新興職位與技能需求

• 人物: Andrew Ng (AI教育布道者)
• 時間: Jun 1
• 熱度: 👀 532,362
• 觀察: AI FDEs和AI Engineers因LLM客製化需求而崛起，且預期AI Engineer的職位數量將大幅超越FDEs。
• 意義: 揭示AI產業落地需要大量客製化人才，創造新的工作機會與技能需求，推動AI人才市場持續擴張。

2. 全球AI安全政策加速

• 人物: Greg Brockman (OpenAI 聯創)
• 時間: Jun 3
• 熱度: 👀 178,509
• 觀察: AI安全政策正累積實質動能，美國已透過行政命令邁出一步，並積極倡議未來監管方向。
• 意義: 預示AI技術的監管框架將加速形成，直接影響AI產品開發、部署與商業化的合規成本與策略。

3. AI藥物發現能力大進展

• 人物: Greg Brockman (OpenAI 聯創)
• 時間: Jun 3
• 熱度: 👀 131,416
• 觀察: OpenAI對GPT-Rosalind進行重大升級，顯著提升其在藥物發現、分析、設計及實驗流程中的智慧能力。
• 意義: 顯示AI在生醫研發領域的應用潛力深化，有望加速新藥上市週期，為生物科技投資帶來新動能。

4. 開源輕量AI模型推廣

• 人物: Jeff Dean (Google DeepMind 首席科學家)
• 時間: 20h
• 熱度: 👀 41,865
• 觀察: Google DeepMind發布Gemma 4 12B，一款功能強大且能直接在筆記型電腦上運行的開源模型。
• 意義: 大幅降低AI模型開發與部署門檻，加速邊緣AI應用普及，促進更廣泛的商業化與創新生態。

5. AI客服安全漏洞警示

• 人物: Pierre Levy (認知科學家)
• 時間: Jun 2
• 熱度: 👀 584,505
• 觀察: Meta AI聊天機器人被發現嚴重安全漏洞，允許駭客透過其修改高權限帳號的敏感資訊。
• 意義: 強調AI代理在接觸真實系統前，必須建立嚴格安全機制與監管框架，以避免巨大的資安與商業風險。

原始動態

其他 | Peter Steinberger

• 時間: Jun 3
• 熱度: 👀 646,192
• 原文: Dear GitHub Copilot team, I am happy to announce that I successfully burned all of my monthly tokens in under 3 days thanks to your garbage new pricing model. I'd also like to inform you that I won't be renewing my subscription or adding more budget. Best, A former customer.

其他 | Elon Musk

• 時間: 7h
• 熱度: 👀 815,898
• 原文: Right-wing extremism is so uncommon that left-wing extremists have to fund it.

其他 | Nassim Taleb

• 時間: 8h
• 熱度: 👀 216,578
• 原文: "For Aristotle, the banausos is not just physically deformed by his profession, but politically and ethically stunted: specialization makes him less a complete man than a TOOL of his occupation. Politics, book VII: \"And any occupation, art, or science, which makes the body or soul or mind of the freeman less fit for the practice or exercise of virtue, is vulgar; wherefore we call those arts vulgar which tend to DEFORM the body, and likewise all paid employments, for they absorb and degrade the mind.\""

重點 | Andrew Ng

• 時間: Jun 1
• 熱度: 👀 532,362
• 原文: "One of the new, buzzy jobs in Silicon Valley is the AI Forward Deployed Engineer (FDE), an engineer who is embedded within a client organization to help customize solutions, such as building and tuning agentic workflows that suit the client’s particular needs. I’ve heard from people who are wondering anew about the FDE career path since OpenAI and Anthropic started building new teams to place FDEs within client organizations. The rise of FDEs for AI workloads is one way AI is creating new jobs (and why the jobpolcalypse narrative of upcoming job market collapse is false -- there will be many AI and non-AI jobs). However, I believe there will be far more AI Engineer jobs than FDEs, as I explain below. The FDE role was pioneered about two decades ago by Palantir, which sent engineers to government locations to work on secure, air-gapped networks. In addition to having good technical skills, FDEs need communication skills and sometimes business skills. For example, they may need to speak with clients to understand their needs, formulate a strategy to prioritize projects, explain complex technology, and respectfully push back if a client asks for something unrealistic. They’re enjoying a resurgence because of the amount of work involved in taking an off-the-shelf LLM and building it into a custom agentic workflow that fits particular business needs. However, I believe the number of AI Engineer jobs will be far larger. A company might accept a few FDEs to be embedded within its organization. But most companies will want far more of their own employees working on their projects. While my organizations do hire FDEs, we hire far more AI Engineers! Also, a common client concern is that it is hard to find vendor-neutral FDEs — they are, after all, there to deeply integrate a particular vendor’s product into a company. In this moment when it’s hard to predict which AI service will be the best one in a year’s time, optionality (the ability to pick whatever vendor turns out to fit best in the future) is very valuable. In contrast, letting FDEs tightly bind a company’s processes significantly reduces optionality. Right now, I see surging demand for AI Engineers who can build software applications using AI software components (like LLM prompting, agentic frameworks, evals, etc.) and effectively use AI coding agents (like Claude Code, Codex, Antigravity CLI, and OpenCode). As the AI Engineer role matures, I expect it to fragment into more specialized roles, like the generic Software Engineer role from decades ago fragmented into frontend, backend, mobile, data engineering, devops, and so on. What will be the future, specialized AI engineering roles? I don’t know. Perhaps there will be AI FDEs, LLMOps Engineers, Evals Engineers, AI Data Engineers, Harness Engineers, and other roles we don’t have names for yet. But for now, I see a lot of AI engineers who are generalists create a lot of value. Skilled AI Engineers are in very high demand! As our field continues to mature over the coming decade, I look forward to new specializations within AI Engineering that create even more job opportunities. [Original text: The Batch newsletter]"

重點 | Greg Brockman

• 時間: Jun 3
• 熱度: 👀 178,509
• 原文: There’s real momentum right now for AI safety policy. Yesterday’s EO on cyber was an important step forward. We’re proposing a set of ideas for policymakers to consider next and to put the US out in front on frontier safety.

重點 | Greg Brockman

• 時間: Jun 3
• 熱度: 👀 131,416
• 原文: "Major upgrade to GPT-Rosalind, with much better intelligence for drug discovery, analysis, design, and experimental workflows:"

其他 | Elon Musk

• 時間: 29m
• 熱度: 👀 694,052
• 原文: Yup

其他 | Nassim Taleb

• 時間: 6h
• 熱度: 👀 88,997
• 原文: As with fat tails Llms are frequency machines that fail to extrapolate outside the sample set. What they know is the VISIBLE. Almost as bad as economists, almost worse than psychologists.

重點 | Jeff Dean

• 時間: 20h
• 熱度: 👀 41,865
• 原文: "Check out our Gemma 4 12B model: it's a super capable open weights model that can run directly on your laptop."

其他 | Nassim Taleb

• 時間: 3h
• 熱度: 👀 41,256
• 原文: When they don't say who attacked, you know it is (what is currently known as) the state of Israel.

其他 | Eric Topol

• 時間: 7h
• 熱度: 👀 7,449
• 原文: The new report builds on our enhanced understand of tumor evolution and promotion. "Promolytics" "\"An ambitious goal would be to identify promotion-preventive agents—promolytics—that could directly inhibit such potentially dangerous clones, or alter the competitive clonal landscape in favour of clones that rarely or never progress to cancers.\"" nature.com

重點 | Pierre Levy

• 時間: Jun 2
• 熱度: 👀 584,505
• 原文: "This is disaster from Meta AI. Imagine being able to hack high profile accounts like White House, the U.S. Space Force, and Sephora simply by chatting with a support bot. why would an AI chatbot be allowed to perform critical actions like changing the email address of an account in the first place? Password resets, email changes, and account recovery are some of the most sensitive security operations. Handing that authority to a chatbot without robust verification seems like a disaster waiting to happen. This is exactly why AI agents should have guardrails before they get access to real systems. Source : 404"

其他 | Huanusa

• 時間: 6h
• 熱度: 👀 6,292
• 原文: 对普通投资者影响最大的一条新规来了！ 6月1日刚刚发布的《国务院关于对外投资的规定》 7月1号就要正式实施了 简单说，这次管得更严了： 哪些地方不能投、哪些地方要报备，都写得清清楚楚 没按规定备案、乱投禁止领域、或者材料造假的，都要处罚 不光罚你本人，管理人员也要连带负责，最高能罚10万 这到底是什么意思，看懂的说说看 尤其是炒美股、炒币、买海外房产、投境外基金的朋友 赶紧看看吧，官方链接放在评论区👇

其他 | Huanusa

• 時間: 8h
• 熱度: 👀 3,525
• 原文: 辛普森一家又“预言中”2026了？ 网友盘点了其中10个最惊人的剧情…… 包括： 比特币大幅崩盘 美国陷入大规模冲突 新型疫情出现 ……等等 最近比特币波动 + 近期一些公共卫生事件， 让不少人又把这些老梗翻出来讨论。 加上之前朱迪、Boma、帕克等人的比特币相关观点， 已经是第四个“指向”了... 你怎么看这些“预言”？

其他 | Fei-Fei Li

• 時間: Jun 3
• 熱度: 👀 5,562,327
• 原文: "World Labs CEO Dr. Fei-Fei Li: \"The world is not made of words.\" \"Language models have given machines an extraordinary command of concepts, vocabulary, and reasoning, but the physical world, virtual or real, runs on a different substrate.\" \"Where language models learn the statistical structure of text, world models learn the statistical structure of space and time: how light falls on a surface, how a garden looks from an angle no camera has captured, how objects respond to force and follow the laws of physics.\" \"Language gave machines a way to talk about that world. World models are how machines will finally come to understand, imagine, reason and interact with it.\" Full piece:"

其他 | Greg Brockman

• 時間: Jun 3
• 熱度: 👀 106,148
• 原文: fly with codex

其他 | Peter Steinberger

• 時間: 17h
• 熱度: 👀 63,336
• 原文: "Here’s the video of my talk at MS Build: Build the thing that builds the thing."

其他 | Eric Topol

• 時間: 4h
• 熱度: 👀 16,150
• 原文: Home alone. Remote work can adversely influence mental health. A new study

其他 | Jeff Dean

• 時間: Jun 3
• 熱度: 👀 78,387
• 原文: This is great to see. The scientific community & the general public should be in the habit of celebrating amazing scientific achievements with standing ovations 🎉

其他 | Yann LeCun

• 時間: Jun 3
• 熱度: 👀 71,676
• 原文: "Can MLLMs actually track what's happening in a video? Introducing VSTAT 🎯, our new benchmark for visual state tracking. The tasks are simple: count cups, read typed words, count page flips. Humans solve them easily. MLLMs don't." 🧵 [1/11]

其他 | Jeff Dean

• 時間: Jun 1
• 熱度: 👀 31,543
• 原文: It was really fun to chat with the legendary . I asked him about things that only he may know the answer to. 🙂 Link to full video in thread.

其他 | Eric Topol

• 時間: 8h
• 熱度: 👀 9,896
• 原文: "\"Administration figures are working a pathway to regulate independent AI doctors.\" by" gift link

其他 | Eric Topol

• 時間: 7h
• 熱度: 👀 6,285
• 原文: If helpful to understand the importance of this study, I made an infographic with

其他 | Huanusa

• 時間: 6h
• 熱度: 👀 610
• 原文: 这事影响很大，富途都发邮件说明了

其他 | Greg Brockman

• 時間: Jun 3
• 熱度: 👀 3,261,198
• 原文: It's time to fly.

其他 | Sean Kelly

• 時間: 5h
• 熱度: 👀 17,357
• 原文: Most people do not understand that the real AI wall is the rapidly increasing energy need

其他 | Sean Kelly

• 時間: Jun 2
• 熱度: 👀 6,652,654
• 原文: I will soon be introducing a bill to give the public a 50% ownership stake in the largest AI companies in America. This would guarantee that the trillions created by AI are used to improve the lives of all of us — and block oligarch decisions that harm the American people.

其他 | Thomas Wolf

• 時間: Jun 3
• 熱度: 👀 14,595
• 原文: 👀

其他 | Sean Kelly

• 時間: Jun 3
• 熱度: 👀 21,017
• 原文: Astrophysicists have found a Giant Ring in space that should not exist youtube.com

其他 | Pierre Levy

• 時間: Jun 3
• 熱度: 👀 8,388
• 原文: Good AI alignment is necessarily EA adjacent. But EA failed to derive a global morality, not for lack of good intentions or intelligence, but because human rationality is a brittle tool, even if you are a good rationalist, and most people prefer to either muddle through or to consult prophets and leaders. Obviously that won't suffice for powerful AI.

其他 | Sean Kelly

• 時間: Jun 3
• 熱度: 👀 34,078
• 原文: Ok, somewhat more seriously, has been going on about the need to change the social contract for some years, and he is right, of course. If AI reaches human level intelligence at a labor cost below that of most current professions, then a lot of people will lose purchasing power extremely quickly. You don't need a degree in economics to see that this is not a stable situation. Though I recently read a 50 page paper with an elaborate economics model that, well, arrived at this very conclusion. The obvious thing to happen in that situation is that governments will nationalize AI companies. Sanders is somewhat early to the game, but then again who knows if he'll live to see AGI, so it kinda makes sense, or at least not less sense than American politics in general. That said, a group from OpenAI proposed already in April to have a "public wealth fund" to give people (presumably US citizens) shares in the AI-generated wealth (see below). I guess that's the smart thing to say if you live in a country where the average Joe is armed to the teeth and drives a bulletproof car, whether or not you actually mean it.