tnwu16
即係我覺得而家你，其實 Anthropic 同埋 OpenAI 嗰個建成格局係咩呢？就係一個佔領咗嗰個山頭，即係佔領咗山峰呀。即係一個可以遙望晒所有嘅，即係遙望晒所有低地，即係睇晒所有嘅。即係不如其實佢，睇下先呀。係我繼續講，即係嗰度你當初賺你 OpenAI 靠用，即係錢。即係本來你佔領嘅最大互聯網入口就係 Google 嘛，即係 Alphabet 嘅 Google Search 嘛。即係佔領咗個山頭嘛，但係你 OpenAI 通過個技術嘅，即係各位打擊或者新嘅技術嘅，即係例如你本來佢佔領，但係你用戰略分炸機去炸平咗佢嗰個 Google Search 嗰個佔領嘅山峰，即係嗰啲山地，即係最大嘅覆蓋面積嗰度做咩呢？即係就去搶奪咗呢個消費者入口嘛。即係等於你嗰個最高流量就俾 OpenAI 了嘛，係咪先？即係我講到呢度先。但問題就係 Anthropic 佢嗰做，你……佢就即係等於佢佔領咗而家就係佢屋企，依家 OpenAI 佔領咗，等於喺呢個 OpenAI 佔領嗰個山體嗰度呢，即係個山嘅表面，山峰嗰度呢，就係喺下面挖咗條地道落去，就係佢做嘅。即係佢挖咗佢好似香港一隧道或者個獅山隧道，即係貫穿咗成個山體，作為一個好似哥倫山嗰啲軍事基地咁樣，係即係等於係一個掩體呀。即係佢嗰個 Anthropic 佢個位置就係處於在佢係叫二守難攻。即係佢好難，你好易守嘅，但係進攻 OpenAI 係唔係咁容易，因為佢打 OpenAI 消費者 market 唔係咁容易。但係 OpenAI 就係佢，其實已經打殘咗 Google Search 啦，我已經接近。即係基本上而家可以勝負差唔多係已定嘅。就算佢有得，基本上係佢只係剩返啲……即係佢佔領咗最高，即係等於最高峰嗰個山峰喺度佔領咗佢嗰啲叫做其他山腰嗰啲南山，都係可以通過時間慢慢去磨平佢 Google Search 嘅優勢啦。但係其實 OpenAI 佢嘅危險就係佢打 Anthropic，其實佢唔會打得打 Google Search。即係其實打 Google Search 都打得好痛苦，坦白講，都唔係容易打，都係一場苦戰。即係都係一場消耗戰呀，你有辦法可以用資金嘅優勢去打贏佢嘛，係咪先？即係你可以用技術同埋用資金優勢去雙重打擊佢贏佢嘛。但係你如果你要打咩呢？OpenAI 你打 Anthropic 就會難好多啦。即係因為你唔能夠直接去用……即係用地氈式轟炸去打你，你真係要逐個逐個，即係你慢慢磨呀。即係你會用人力，你用時間成本。即係你唔可以再用叫做算力補貼呀，即係嘅方法去……佢好似打 Google Gemini，打 Google Search 上咁容易去，即係靠慢，用大力出奇蹟嘅方法去贏呀。你變咗係真係要好似做手術咁樣，你要好似即係用鑽笨錐，即係你用火焰噴槍，慢慢去將……即係好似俄羅斯打壓縮鋼鐵廠咁樣，或者以色列打哈馬斯地道戰咁樣，等於你要將 Anthropic 企業端嘅 AI 逐個逐個企業走去盤根錯節咁樣走去慢慢去同佢哋傾。即係你唔能夠用錢去擠出個市佔率出嚟。你唔可以用錢去憑空去擠出一個市佔率，即係擠出一個大窿出嚟，唔得㗎。你係要好似係血肉磨坊，即係好似蘇聯去打呢個納粹嗰陣時個國會咁樣，係要逐間逐間幢清掃。即係等於係一種會做血肉磨坊，即係一種我唔識講咁樣呢。同埋最大問題就係你 OpenAI 佔領嗰度山峰呢，嗰個高地呢係好容易攻，即係攻人哋嘅，但係佢係好難守，因為佢係會成為全世界嘅焦點，
即係全世界都想同OpenAI爭對呢一個消費級嘅應用，
係唔係先？
即係雖然你OpenAI已經喺個山峰度架好曬好多防空嘅炮塔去防止啲人再用戰術轟炸機去炸你啦，
同埋你要守住，
但係即係話其實你可以磨平個山峰，
去起機場，
然後起飛啲戰鬥機，
去即係攔截人哋嗰一啲叫做……
即係準備嚟轟炸你呢一個戰略高地嘅，
即係戰術轟炸機，
又或者叫做，
又或者起飛啲戰術轟炸機去轟炸人哋嗰一啲……
即係其他嗰一啲即係山峰比你低嘅，
即係總之就係冇咁……
即係我話咩嘢呢，
即係冇嗰一個叫做，
即係喺個戰略嘅……
即係點樣講呢，
即係轟炸戰略低地，
即係喺嗰一個叫做，
喺海拔冇你咁高嘅即係戰略高地啦，
咁相對於你嚟講就係低地啦，
明唔明呀，
咁係因為你理論上佔領咗個山峰先呀嘛，
但係你打如果要打山體呢，
其實你個優勢就好難……
因為你冇得用……
即係冇得喺山峰上面起飛呢一個戰略轟炸機去同人哋打空戰，
起飛呢一個戰鬥機去同人哋打dogfight，
唔得嘅，
你一定要呢，
即係你知道依家你OpenAI就係主要佔領咗成個AI market嗰一個山體嘅表面，
Openai就佔領咗山峰上面嗰一啲叫做係少量嘅地道啦，
但係成個山嘅掩體，
即係最底下嗰一個掩體呢，
成條最……
貫穿得成個山體最長嗰一條隧道呢，
即係最底下嗰一條隧道呢，
即係譬如話香港仔隧道、獅子山隧道又或者大老山隧道咁長嗰一條，
又或者長青隧道，
即係最長，
即係由海平面啱啦grow出嚟，
即係最底嗰一條隧道呢，
其實係由Anthropic去佔領嘅，
甚至地下嘅隧道都係由Anthropic佔領，
等於你要慢慢去，
即係等於你要落山去揼佢嘅，
你唔可以利用你嗰一個叫做……
即係你唔可能……
即係openai你佔領咗成個山峰唔代表你可以將嗰一個……
即係你始終你都要落山去同佢入個山洞度同佢慢慢血拼囉，
所以我就覺得你打Anthropic就一定係一……
即係你……
嗰一個都唔係一場叫做曠日持久戰，
佢唔係資本戰，
佢唔能夠……
即係佢唔係閃電戰嚟，
即係你打Google Search，
你可以用閃電戰去打贏佢，
你打Anthropic一定係變成係曠日持久嘅，
即係好似亞速鋼鐵廠嗰一啲咁樣嘅模式囉，
即係嗰一個叫做難以寸進嘅，
即係你唔能夠用優勢火力去碾壓佢哋地毯式轟炸去實施呢一個叫火力壓制，
唔得㗎，
即係你只能夠係一命換一命咁樣，
即係變咗係冇……
你知唔知我講咩嘢？
即係依家我覺得就係Anthropic同OpenAI就係處於咁嘅格局，
即係競爭格局就係一個你OpenAI佔領咗成個山嘅表面，
即係山體嘅表面，
變咗同埋少量山頂嘅地道，
但係你Anthropic係貫穿成個山體最底嗰一層同埋山體下邊嗰一啲掩體，
全部都係Anthropic食曬，
咁你點打呢？
你冇得去……
即係你鑽地彈都鑽唔到咁深㗎嘛，
大佬。
Thought for 23s
正在理解對方用粵語提出的比喻。
確認對方粵語比喻的意思，並準備用粵語回應。
正在比較兩家公司在市場上的實際佔有率。
正在評估這個比喻有幾分站得住。
正在查最新的市場數據，核實雙方的實際處境。
我明你講嘅意思，不過你個格局入面有啲數字係事實問題，我先查下而家（2026年11月）嘅最新情況，唔想憑舊印象同你講。
Sam Altman, chief executive officer of OpenAI
techcrunch.com
Anthropic Tops OpenAI in Business Customer Share
aiweekly.co
Sam Altman, chief executive officer of OpenAI
techcrunch.com
forkast.news
forkast.news
Anthropic Tops OpenAI in LLM Revenue Share
letsdatascience.com
OpenAI surpasses Anthropic in Q3 enterprise growth, 82% to 76%
cryptobriefing.com
Autonomous Agents, Ads, and Infrastructure: This Week in Enterprise AI
azumo.com
OpenAI leads, Anthropic surges as enterprise AI shifts to multi-model reality
www.emarketer.com
LLM Provider Revenue Analysis - December 2025
unmistakablemedia.getoutline.com
openai still leads enterprise ai but anthropic is gaining fast according to new study
the-decoder.com
Anthropic Claude 3.5 logo
techcrunch.com
Global LLM Adoption and Revenue Snapshot, Q1 2026
counterpointresearch.com
Anthropic Gains on Enterprise AI Spending Share
letsdatascience.com
LLM Market Share 2026: ChatGPT, Gemini, Claude and More
www.usecarly.com
Two Companies, $44 Billion, and the Future of Enterprise AI
algeriatech.news
research llm provider market share 2026
enterno.io
Enterprise AI Platforms 2026
neuronfeed.com
Weighing a news item on enterprise market share.
Checking the cited market-share figures against Ramp and Similarweb data.
Failed to fetch: https://cryptobriefing.com/openai-surpasses-anthropic-q3-enterprise-growth/
Failed
October 2026 Ramp AI Index: Cracks in the AI thesis, part 3
ramp.com
Ramp AI Index
ramp.com
Anthropic vs OpenAI Business Adoption in 2026: What the RAMP Data Shows
www.mindstudio.ai
Ramp AI Index: AI business adoption re-accelerates, led by Anthropic
ramp.com
Anthropic finally beat OpenAI in business AI adoption — but 3 big threats could erase its lead
venturebeat.com
Anthropic Overtakes OpenAI in Business AI Spending
aiweekly.co
Anthropic overtakes OpenAI in B2B adoption for the first time according to Ramp spending data
the-decoder.com
Ramp AI Index Oct 2026: Opus 5.5 "34%" Claim Checked
explainx.ai
Anthropic Overtakes OpenAI in U.S. Business Adoption for the First Time, Ramp Data Shows
www.bitsminds.com
How did Anthropic do it?
econlab.substack.com
Reviewing the newest market-share data.
Comparing growth signals across businesses using multiple models.
Weighing conflicting share data across two different metrics.
Checking Similarweb and Google Search claims against the cited figures.
Alphabet Inc. - Form 8-K - FY2026
www.sec.gov
9to5google.com
9to5google.com
alphabet googl q3 2026 earnings call transcript
singjupost.com
Alphabet Q2 FY 2026: Google Cloud Leads Growth Amid Rising AI Investment
futurumgroup.com
Alphabet Q2 FY 2026: Google Cloud Leads Growth Amid Rising AI Investment
futurumgroup.com
www.mexc.com
www.mexc.com
Alphabet’s 2026 capex forecast soars; Gemini 3 AI model is a huge success
techblog.comsoc.org
Alphabet's revenue crosses $100 billion for 1st time, Gemini tops 650 mn monthly active users
ianslive.in
google stock ai gemini why bofa bank america still buying
decrypt.co
גוגל מדווחת על 750 מיליון משתמשים ב-Gemini והשקעות ענק ב-AI
www.gadgety.co.il
Checking the user’s claim against Alphabet’s latest search revenue figures.
Weighing Google’s growth against the claim it has lost.
OpenAI Statistics 2026: Revenue, Users & Funding
www.gradually.ai
OpenAI’s Billion Users, US$852bn Valuation and Stargate
aimagazine.com
ChatGPT Growth Timeline
blog.mean.ceo
OpenAI Statistics 2026: Users, Revenue & Market Share
www.getpanto.ai
ChatGPT
en.wikipedia.org
OpenAI ChatGPT Statistics 2026: The Definitive Data
sociallyin.com
ChatGPT Statistics 2026: Users, Revenue, and Enterprise Adoption - Second Talent
www.secondtalent.com
ChatGPT Statistics Q3 2026: 1.2 Billion Weekly Users - TechnologyChecker.io
technologychecker.io
ChatGPT Statistics 2026: 1.2 Billion Weekly Users, Sourced
whizi.io
ChatGPT Statistics (2026): 60+ Data Points on Users, Revenue, and Growth
arvow.com
Judging source quality for the figures.
Comparing revenue sources from several reports.
Fetched: Anthropic finally beat OpenAI in business AI adoption — but 3 big threats could erase its lead
https://venturebeat.com/technology/anthropic-finally-beat-openai-in-business-ai-adoption-but-3-big-threats-could-erase-its-lead
Planning counterpoints against the metaphor.
Weighing how easily users could switch between models.
Weighing compute limits and costs against the metaphor.
Comparing shares of enterprise spending across the AI providers.
- Access the original Anthropic blog from here
www.medianama.com
www.webull.com.au
www.webull.com.au
Enterprise DNA
enterprisedna.co - Cryptopolitan
cryptopolitan.beehiiv.com
Anthropic S-1: 85% Margins, a $265B Compute Bill, and What It Means for Claude Builders
www.buildmvpfast.com
7 min read
beam.ai
cbonds.hk
cbonds.hk
Anthropic announces deal with Google, Broadcom, says revenue has tripled
finance.yahoo.com
October 27, 2026·57 min read
sec-api.io
how anthropic and google plan to challenge nvidias ai dominance
ioplus.nl
futuresearch.ai
futuresearch.ai
Anthropic Financial Growth and OpenAI Concerns
surfboard.bandprotocol.com
anthropic closes in on a trillion dollar valuation
www.techmarketview.com
ai news today may 22 2026
www.buildfastwithai.com
google 40 billion anthropic tpu investment 2026
www.noqta.tn
anthropic series h enterprise growth analysis
www.claudeainews.com
SpaceX Could Hit $60B Revenue Run Rate by End of 2026
www.basenor.com
⚡ Key Takeaways
algeriatech.news
sid 9ScKC0AhZg
www.longtermwiki.com
Reviewing the funding and revenue figures.
Checking how the revenue claims were sourced.
Weighing the reported revenue figures against each other.
Weighing how the two firms’ tunnel and peak positions compare.
我哋而家評緊OpenAI點樣喺企業市場同Anthropic競爭。
我哋而家諗緊點樣喺企業市場搶佔份額。
我哋而家檢討呢個市場嘅整體格局同日後走向。
我哋而家比較緊Anthropic同OpenAI嘅攻守策略。
我哋而家諗緊點樣整理結論同收尾。
我哋而家核對緊企業市場嘅付費數據。
我哋而家評緊OpenAI同Anthropic喺企業市場嘅攻守消耗。
我哋而家檢視兩間公司喺企業市場嘅付費份額同策略。
我哋而家核對緊Google最近嘅Q3財報同Search增長數字。
我哋而家核對緊Google Search仲有冇持續增長嘅證據。
我哋而家核對緊各方對兩間公司收入同市場地位嘅說法。
我哋而家比較緊兩間公司嘅數據同市場策略。
我哋而家整理緊對方嘅論點同比較結論。
我哋而家檢視兩間公司喺企業市場嘅攻守策略。
我哋而家整理緊OpenAI點樣應對Anthropic嘅策略。
我哋而家比較緊Google同OpenAI嘅流量同付費數據。
我哋而家核對緊Anthropic嘅服務穩定性同算力限制嘅說法。
我哋而家核對緊年中報道嘅時間同來源可信度。
我哋而家核對緊Ramp同Similarweb嘅數據來源同樣本範圍。
我哋而家比較緊Google Search同Anthropic喺企業市場嘅定位。
我哋而家核對緊企業市場嘅策略同市場動態。
我哋而家整理緊各方嘅論點同市場判斷。
我哋而家整理緊對兩方策略嘅判斷同理據。
我哋而家比較緊Google Search同OpenAI喺流量上嘅攻守情況。
我哋而家核對緊隧道比喻嘅前提同客戶流動性。
我哋而家比較緊OpenAI點樣應對Anthropic嘅攻守方法。
我哋而家整理緊各方對企業市場攻守格局嘅判斷。
我哋而家比較緊Google Search同Anthropic喺企業市場嘅定位。
我哋而家諗緊OpenAI點樣喺企業市場用三種策略同Anthropic較量。
我哋而家諗緊OpenAI點樣用三招同Anthropic較量。
明，你個比喻我跟到。OpenAI 佔咗山面同山頂（消費者入口、最大流量），連 Google Search 嗰個山頭都差唔多畀佢轟到定局。Anthropic 就喺山腹挖咗最長最深嗰條隧道，即係企業客戶同佢哋嘅工作流程，易守難攻。OpenAI 嘅「空軍」（資金、算力補貼）轟唔到落去，只可以逐間公司、逐個 workflow 咁磨，所以係消耗戰，唔係閃電戰。
先聲明：我係 Anthropic 造嘅，呢個題目我難免有偏向，所以我啱啱查咗最新數據，連對 Anthropic 唔利嘅部分都會講。大方向我同意一半：山頂難守、企業市場係消耗戰，都有數據撐，但有兩個前提我覺得唔啱。
數據撐到你嘅部分。 山頂真係難守：據 Similarweb，ChatGPT 佔全球生成式 AI 網站流量，由一年前 73.3% 跌到今年 8 月 55.5%，Gemini 已經有 25.6%。企業方面睇 Ramp（佢睇美國公司用信用卡同賬單畀咗幾多錢，樣本未必代表大型企業），會見到「深」同「闊」嘅分別：以有畀錢嘅公司數目計，9 月 Anthropic 係 43.8%、OpenAI 係 40.6%，而且 OpenAI 嘅升幅大過 Anthropic。但以 token 開支計，10 月 23 日嗰個禮拜 Anthropic 佔 60.5%，OpenAI 由 44.2% 跌到 35.1%。即係 OpenAI 勝在闊，Anthropic 勝在深，同你山面對隧道嘅講法幾貼。
第一個唔同意：Google Search 冇被轟平。 Alphabet 最新 Q3，Search 收入升 12%，查詢增長率係五年嚟最高，Cloud 升 103%，而且Gemini app 已經有 10 億用戶。Google 唔係被炸平嘅山頭，而係喺山頂同 OpenAI 對轟，仲有自己嘅 TPU 晶片同豐厚利潤做後盾。所以 OpenAI 唔係「山頂搞掂咗，剩低對付 Anthropic」，而係上面有 Google、下面有 Anthropic，兩面受敵。呢點對你個結論其實加分，但「Google 差唔多完」呢個前提要收返。
第二個唔同意：隧道嘅住客隨時可以搬。 軍事上隧道難攻，係因為守方有地形保障。但 LLM 客戶轉模型嘅成本好低，好多公司同時買多過一個模型，而且Ramp 首席經濟學家形容，從未見過咁動態嘅軟件行業，新對手幾個月就可以推翻領先者。OpenAI 一個月跌 9 點，一方面顯示 Anthropic 動能勁，另一方面都證明呢啲份額係潮水，唔係岩石，潮水可以反方向流。
Anthropic 自己都有漏位（以下多數係年中報道，之後可能已有改善）：增長好大部分靠 Claude Code 一隻產品；曾經頻繁 outage 同 rate limit，仲因為算力唔夠要拒絕部分生意；Uber 四個月就用晒全年 AI 預算，同時試緊 Codex 做後備，而 Codex 水準相近但平啲。另外Ramp 指出，OpenAI 同 Anthropic 80% 嘅開支來自 1% 嘅公司，即係隧道嘅住客其實係少數大戶。
咁 OpenAI 點打？ 鑽地彈鑽唔到，但隧道一樣要補給，所以我覺得佢唔會正面衝，而係三招：
- 斷補給： Anthropic 嘅弱點係算力同成本，OpenAI 三月融咗 1,220 億美元，可以向受夠 rate limit 同賬單嘅大客戶提供更多 capacity、更平嘅價。
- 打咽喉： Anthropic 嘅企業優勢好大程度靠 coding，Codex 同 Claude Code 呢場仗一旦倒轉，隧道最值錢嗰條動脈就出事。
- 山頂餵隧道： OpenAI 話 ChatGPT 每週觸及 12 億人，員工自己帶返公司用，再由下而上變成企業合約。
事實上 OpenAI 都挖緊自己嘅隧道：財務長 8 月已經話企業收入追過消費者，企業客戶超過 200 萬。
而且 OpenAI 唔使贏晒，企業可以兩邊用，佢只需要分到一份。真正嘅決戰係模型水準最後拉唔拉近：拉近嘅話，掌握用戶入口嗰個佔優；拉唔近（尤其係 agent 同 coding），掌握最強模型嗰個佔優。所以我會話呢個市場唔係石山，係沙丘，每次有新 model，風一吹，地形就變。隧道有價值，但唔係堡壘。