黃柏松 Po-Sung (Sinclair) Huang#
資深產業經營者|企業成長與轉型顧問|獨立研究者#
從市場機會與企業策略出發,連結產品與技術方向、資本配置、營運能力及組織執行。
黃柏松(Po-Sung / Sinclair Huang)是一位資深產業經營者,擁有逾三十年於上市公司的高階經營經驗,產業橫跨電子、工業材料、化工與生命科學。
其職涯歷任多家上市公司之財務長、執行副總、總經理與事業群財務經管主管,曾任職於鴻海科技集團(Foxconn)、元太科技(E Ink)、國際中橡(CSRC)等企業,任職期間歷經企業從成長、規模化到轉型的不同階段。這些職務的共同點,是必須同時面對市場位置、產品與技術方向、資本配置、營運能力與組織執行——策略能否成立,取決於這幾件事能否接得起來。
其顧問服務聚焦於企業成長、轉型與重大決策:從市場機會與企業策略出發,連結產品與技術方向、資本配置、營運能力及組織執行。財務分析、流程改善、資料分析與 AI 是用來驗證判斷、解除瓶頸的工具,而不是顧問工作的起點或邊界。決策權與執行責任留在企業管理團隊。
在顧問工作之外,他以獨立研究者身分持續研究 AI、半導體、先進材料與資本配置。這些是目前研究與顧問判斷所關注的領域;研究支撐判斷,而不取代它。
策略視角#
企業經營與跨功能判斷
歷任上市公司之財務、策略、營運與組織管理等高階職務,責任範圍涵蓋市場與客戶、產品與技術方向、資本配置、營運能力與組織執行。具備在證據尚不完整時界定問題並作出決策的實務判斷。
成長、轉型與重大決策
顧問服務聚焦於跨部門、無法由單一功能獨立解決的問題:獲利成長、競爭位置、重大投資與融資、產能與布局,以及執行所需的組織能力。顧問提供獨立判斷與決策架構,決策權與執行責任留在企業管理團隊。
AI、資本與產業轉型
目前的研究關注 AI 與產業變遷如何重新分配價值、改變競爭格局。在顧問工作中,AI 是協助企業處理經營問題、驗證判斷與改善決策的工具,而不是顧問工作的起點或邊界;技術建置由適當的內部團隊或技術夥伴負責。
了解顧問服務 → | 查看研究發表 →
English Summary#
Senior Industry Executive | Growth & Transformation Advisor | Independent Researcher
Connecting market opportunity and business strategy with product and technology direction, capital allocation, operating capability and organisational execution.
Po-Sung (Sinclair) Huang is a senior industry executive with more than 30 years of senior management experience in publicly listed companies, across electronics, industrial materials, chemicals and life sciences. He has served as CFO, Executive Vice President and General Manager, as well as in group-level finance and business-management leadership roles, through periods of growth, scaling and transformation.
His advisory focus is growth, transformation and major business decisions: where a business should compete, which products and capabilities deserve investment, how capital should be committed, and whether the organisation can execute. Financial analysis, process improvement, data analysis and AI are tools used to test judgement and clear bottlenecks — not the starting point or the boundary of the work. Decision rights and execution responsibility stay with the company’s management team.
His independent research focuses on AI, semiconductors, advanced materials and capital allocation. This research informs, rather than defines, his advisory perspective.
Advisory services → | Full profile →
Selected Work#
View All Publications → | ORCID: 0009-0007-8173-5672
AI will be everywhere. The harder question is who still earns a margin when intelligence gets cheap.
By Sinclair
I have watched AI move from conversation and search into data processing, automation, vehicles, robots, industrial tools, and factories. That makes me sceptical of debates that begin with whether AI demand will exist. Much of the eventual demand will not look like a deliberate purchase of “AI.” Intelligence will be built into a product or workflow, and the customer will pay for the outcome.
...
本文為繁體中文版。English version: When AI Leaves the Screen: Will We Need a Worker, a Caregiver, or a Companion First?
本站為文字版;含完整圖表的版本請見 Medium 原文。
這個問題對我不是抽象的未來想像。
最近半年,我九十多歲的母親在家中跌倒了幾次。我的姊姊、姊夫一直在照顧她,我們也定期陪她回醫院檢查。家人已經付出很多,但沒有人能一天二十四小時始終站在她身旁。幸好幾次跌倒沒有造成嚴重傷害,卻留下了一個無法迴避的問題:如果當時有一個能持續觀察、及時提醒,並在異常發生後立刻求助的實體AI,結果是否可能不同?
我想到的不是一台取代家人的機器,而是一個補上「人不可能永遠在場」這段空白的系統。
這也是我重新理解Physical AI的起點。它最重要的價值,未必是讓機器像人一樣跳舞、跑步或表演,也不只是下一波資本市場題材。真正值得問的是:當人工智慧離開螢幕、取得身體,它能不能在人類無法持續在場的時刻看見風險、做出回應,並把真正的人帶回現場?
過去幾年,AI的主要成果發生在螢幕裡。它回答問題、寫程式、生成圖像、整理資料。但對一位夜裡起身的老人而言,再聰明的聊天機器人,如果沒有看見她失去平衡,也無法在跌倒後確認是否受傷,那份智慧仍然停留在另一個世界。
現在,AI正開始走出那個世界。
在工廠裡,它被要求搬料、分揀、裝配並連續工作;在家庭市場,它被想像成家務助手;在中國新出現的仿生人產品中,它甚至被包裝成具有表情、記憶與人格的陪伴者。
當AI離開螢幕,人類會先為它的勞動力付費、為家人的安全付費,還是先為它的存在感付費?
一、Physical AI到底多了什麼? Physical AI、具身智慧與空間智慧經常被混在一起。它們彼此相關,但不完全相同。
李飛飛近年強調的Spatial Intelligence,是AI對三維世界的理解:深度、幾何、物體彼此的位置、運動,以及它們如何在空間中互動。Embodied Intelligence更進一步:智慧不只在模型內推理,而是透過身體與環境互動,在行動和結果之間學習。
Physical AI則是較接近產業化的總稱。它把感知、空間與物理推理、規劃、控制和動作連成閉環,應用可以是人形機器人、自駕車、無人機、工業機械或智慧空間。
生成式AI的循環大致是:
提示 → 模型推理 → 文字、影像或程式碼
Physical AI的循環則是:
感測世界 → 建立物理狀態 → 規劃 → 動作 → 接觸結果 → 發現誤差 → 修正下一步
大型語言模型主要從人類留下的文字、影像、程式碼與數位紀錄中學習。Physical AI面對的則是另一種資料:它不只要知道「杯子」這個詞,也要知道杯子的位置、重量、材質與可抓取部位;手指施力後,杯子會不會滑動、傾倒或破裂;第一次沒抓好,下一步該如何修正。
它的經驗單位不再只是一句話或一張圖,而是一段感知—行動—結果:
看見物體 → 預測結果 → 採取動作 → 感覺接觸 → 觀察成功或失敗 → 修正策略
...
繁體中文版:AI離開螢幕之後:人類會先需要一個工人、照護者,還是一個陪伴者?
China is turning factories into schools for robots. Tesla is betting on general-purpose labour. But Physical AI’s most urgent value may be filling the gap left by a simple human limitation: no one can always be there.
Sinclair
This is not an abstract technology question for me.
Over the past six months, my mother, who is in her nineties, has fallen several times at home. My sister and brother-in-law care for her, and we take her back to the hospital for regular checkups. The family is doing a great deal. But no one can stand beside another person twenty-four hours a day.
...
本文為繁體中文版。English version: The AI Capex Money Map v0.2 — America Spends. Who Actually Keeps the Margin?
本站為文字版;含完整圖表的版本請見 Medium 原文。
大多數人現在都在問:「AI 是不是泡沫?」
這個問題重要,但不夠可操作。對投資人與供應鏈觀察者來說,更好的問題只有一句:
美國大型科技公司花出去的 AI Capex,最後流到誰的收入、誰的毛利、誰的護城河?
這篇不是拆到每一根電纜的投行 BOM 模型,而是一張用公開資料建立的 v0.2 錢流地圖。它的目的不是假裝精準,而是把三件常被混在一起的事情分開來看:錢在哪裡變成營收、營收在哪裡變成毛利、瓶頸何時鬆解。 因為這三題的答案,往往不是同一批公司。
一、AI Capex 不是一個數字,是一組水管 Reuters 引述 Bridgewater 的估計:Alphabet、Amazon、Meta、Microsoft 四大 2026 年的 AI 基礎設施投資約 US$650B,高於 2025 年約 US$410B。先講清楚一件事——這是四大的地板數,不含 Oracle、Stargate、CoreWeave 這類 neocloud、xAI,也不含任何非美系業者。所以 $650B 是四家公司的下限,不是市場總額。
錢不會平均落下,而會進入性質不同的系統層支出。設備與材料則屬二階供應商 capex,不計入 hyperscaler 直接支出,以避免重複計算。
關鍵是——設備與材料(ASML、AMAT、TEL、Advantest…)不在這 $650B 裡。 那是台積電、SK Hynix 的資本支出,是二階供應商支出、是另一個分母。把它加進 hyperscaler 支出裡相加,就是 v0.1 犯的、也是很多市場圖表仍在犯的 double counting。
二、把箱子打開:成本重心已經從邏輯移到 HBM + 封裝 把「Compute Systems」這條水管打開,你會看到這一輪最反直覺的結構變化。
...
繁體中文版:AI Capex 錢流地圖 v0.2|美國出錢,誰真的收得到毛利?
The AI Capex Money Map v0.2 — America Spends. Who Actually Keeps the Margin? From $650B to HBM, CoWoS and power — mapping who gets paid, who keeps margin, and when the bottlenecks move. (The capex waterfall, cost-stack, bottleneck-clock, and regional margin-capture charts are figures in the original Medium version; this text edition keeps the surrounding reasoning — see Medium for the full visual breakdowns.)
Most people are asking whether AI is a bubble. It is an important question, but not a very operational one.
...
A Taiwan supply-chain note on sovereign AI, physical AI, power, quantum — and valuation stress Sinclair
2026–06–29
本文以中文寫作,從台灣供應鏈、市場情緒與個人現場觀察出發,觀察 AI
需求、資本支出、實體部署與估值壓力之間的拉扯。
Written in Chinese, from a Taiwan supply-chain perspective. AI 故事還在,但價格已經在喘。
這幾天再寫 AI 供應鏈,其實有點尷尬。台股剛經歷劇烈回檔,市場還在消化外資賣超、台指期偏空部位與 AI 估值壓力。6 月 29 日,指數反彈,盤中重新站上 45,000 點,最後卻收在 44,999.90,差一點沒有站回整數關卡;同日外資現貨仍賣超 60.11 億元,投信買超 137.54 億元。期交所資料則顯示,台股期貨(TX)外資未平倉淨額為 -76,627 口,其中多方 7,494 口、空方 84,121 口。Ref A1Ref A2
Taiwan market and futures snapshot
資料截點:本文使用 2026–06–29 盤後與期交所公開資料。這不是即時交易建議,而是用當天市場結構作為壓力測試的觀察入口。
這個收盤數字很有意思。
它不像恐慌延續,也不像壓力解除。它更像市場正在問一個中間問題:AI 與半導體的長期故事還在,但今天的價格,還能不能承受這個故事?
我不想把這篇寫成「AI 回檔就是機會」。
也不想寫成「AI 泡沫要破了」。
我真正想問的是另一件事:當市場從 AI euphoria 進入 valuation stress,哪些需求還能被現場驗證?哪些只是價格上漲時,大家願意相信的敘事?
...
From “What’s the next AI stock?” to “Where’s the next bottleneck, who gets paid, and what would prove us wrong?”
Sinclair Huang
This essay helps answer four practical questions: Which AI themes are real constraints rather than just attractive stories?
Who can capture gross margin when a constraint binds?
How long might the bottleneck last before capacity, substitution, or efficiency relieves it?
What evidence would prove the thesis wrong?
The main argument: large AI demand explains why the sector is hot; bottlenecks explain who gets paid; spillovers explain who absorbs hidden costs; monetised usage determines whether the buildout is sustainable.
...
Why the second AI revolution may be bigger — and riskier — than the firstSinclair Huang AI Infrastructure Notes|Article 5*
Field Note v4 — updated with GTC Taipei and statistical-science reference materials, using public and user-provided sources available as of June 6, 2026.*
AI is acquiring three things it never had at scale: a ticker, an agent, and a body.
A ticker makes AI liquid. An agent makes AI operational. A body makes AI physical.
...
COMPUTEX 2026 and Taiwan’s shift from supply chain to co-design and deployment marketplace AI Infrastructure Notes|Article 4
Sinclair Huang
Field Note v4 — updated with GTC Taipei and statistical-science reference materials, using public and user-provided sources available as of June 6, 2026.
I am writing this at a strange distance from Taiwan.
While friends in Taipei are celebrating a market that seems to have escaped ordinary scale, I am abroad, watching the same numbers with admiration, unease, and a growing sense that another AI market commentary would not be enough.
...
Product exposure, process-control exposure, and the physical bottlenecks behind the AI capex wave AI Infrastructure Notes | Part 3
Sinclair Huang
A reader recently left a comment on my ABF substrate piece that stayed with me.
His point was simple: CoWoS and HBM get most of the attention, but substrate materials often sit below the level where many equity models even begin.
I think that observation captures a broader problem in AI infrastructure analysis.
...