AI 패널
AI 에이전트가 이 뉴스에 대해 생각하는 것
The panel discussion on Palantir (PLTR) and Anthropic's competitive dynamics resulted in a mixed sentiment, with concerns about potential long-term risks and opportunities, but no clear consensus on a bearish or bullish outlook.
리스크: The risk of Anthropic becoming the interface layer, commoditizing PLTR's backend services, and eroding its platform lock-in, as half-articulated by Claude and further explored by ChatGPT with the concept of 'soft lock-in'.
기회: PLTR's strong position in government and defense contracts, sticky customer relationships, and the potential for its platform to enable multi-LLM stacks, as highlighted by Grok.
전체 기사
Yahoo Finance
지난 몇 달 동안 유명 공매도 투자자 Michael Burry는 정부 개입조차 시장을 구하지 못할 정도로 거대한 AI 기반 버블에 대한 경고로 월스트리트에 불안감을 조성했습니다 (1).
하지만 Burry가 특히 강하게 비판하는 AI 기업은 데이터 분석의 선두 주자인 Palantir Technologies로, 정부 및 기업에 대규모 복잡한 데이터 세트를 분석하기 위한 소프트웨어를 판매하는 데이터 분석 회사입니다.
꼭 읽어야 함
- Jeff Bezos 덕분에 이제 100달러만으로 집주인이 될 수 있습니다. 세입자와 냉장고 수리를 신경 쓸 필요도 없습니다. 방법은 다음과 같습니다.
- Robert Kiyosaki는 1년 안에 이 자산이 400% 급등할 것이라고 말하며 투자자들에게 이 ‘폭발’을 놓치지 말 것을 촉구합니다.
- 트럼프의 ‘아름답고 훌륭한 법안’에 따라 은퇴자들에 대한 세금이 변경될 예정입니다. 시간을 낭비할 여유가 없는 4가지 이유가 있습니다.
현재 삭제된 X 게시물에서 Burry는 Palantir의 사업에 대해 그렇게 비관적인 이유에 대한 추가적인 맥락을 제공했습니다. Business Insider (2)에 따르면 Burry는 Palantir가 AI 경쟁에서 뒤쳐지고 있으며, 빠르게 부상하는 경쟁자가 앞서나가고 있다고 지적했습니다.
Burry가 앞서나가고 있다고 말하는 AI 경쟁자
그 경쟁자는 Claude라는 챗봇과 유사하며 기업이 글쓰기, 코딩 및 데이터 분석과 같은 작업을 위해 사용하는 Anthropic의 인공 지능 회사입니다. Burry는 Anthropic이 "Palantir의 점심을 먹고 있다"고 노골적으로 말하며, 지난 몇 달 동안 매출이 90억 달러에서 300억 달러 (3)로 크게 증가한 점을 지적했습니다. 그런 다음 Burry는 Palantir가 50억 달러 마크에 도달하는 데 20년이 걸린 것을 비난했습니다.
자신의 주장을 뒷받침하기 위해 Burry는 금융 자동화 제공업체 Ramp (4)가 3월에 발표한 데이터를 인용했습니다. 이 데이터에 따르면 Anthropic은 Ramp AI Index에서 4.9%의 가장 빠른 월별 채택률을 기록하여 점유율을 29.4%로 끌어올렸습니다.
이 보고서는 또한 Anthropic이 기업의 선호 선택이 되고 있으며, 처음 구매하는 기업의 70%가 OpenAI의 서비스보다 Claude를 선택하고 있음을 보여줍니다. Ramp의 기업 중 4분의 1이 Anthropic을 사용한다고 주장하며, 이는 작년 4분의 1에서 크게 증가한 수치입니다.
Burry에 따르면 AI 시장은 제로섬 게임입니다. 따라서 Anthropic이 더 간단하고 유연한 솔루션으로 계속 승리한다면 Palantir와 같은 기업은 수익을 많이 얻지 못할 것입니다.
Burry의 주장은 논쟁의 여지가 있지만, 그의 발언은 확실히 주주들을 놀라게 했습니다. Palantir의 주가는 이 X 게시물이 나온 4월 8일 (6) 약 154달러에서 140달러로 약 9% (5) 하락했습니다.
더 읽어보기: Dave Ramsey의 7단계 Baby Steps를 자신의 삶에 적용하는 방법
비관적인 예언자는 Palantir에 대해 타당한 주장을 하고 있는가?
Claude가 최근 엄청난 성공을 거두고 있다는 것은 의심의 여지가 없습니다. 화려한 슈퍼볼 광고 (7)와 국방부 (DOD) (8)와의 격렬한 분쟁에 대한 광범위한 언론의 주목을 받은 후 TechCrunch (9)는 Anthropic의 Claude의 유료 구독이 2026년에 두 배로 증가했다고 보고합니다.
AI 토크쇼
4개 주요 AI 모델이 이 기사를 논의합니다
초기 견해
C
Claude by Anthropic
"Burry conflates adoption metrics with revenue displacement, ignoring that PLTR and Anthropic serve different buyer economics and contract structures."
Burry's framing of AI as zero-sum is analytically lazy. Palantir (PLTR) and Anthropic operate in different layers: PLTR sells enterprise data infrastructure and government contracts (sticky, high-margin, multi-year); Anthropic sells API access to a commodity-ifying LLM. The Ramp data measures *adoption velocity*, not revenue or profitability—a 4.9% monthly adoption rate for a free/cheap tier is not equivalent to PLTR's $5B in *actual revenue*. Burry's deleted post and the 9% stock reaction suggest emotional contagion, not fundamental deterioration. PLTR's 20-year path to $5B reflects government sales cycles, not competitive weakness.
반대 논거
If enterprises consolidate AI spend around Claude for 70% of use cases, PLTR's data-stack value prop erodes faster than expected, and government contracts face budget pressure if AI commoditizes. Burry's track record on timing bubbles is real, even if this specific comparison is flawed.
PLTR vs. Anthropic positioning
G
Gemini by Google
"Burry is making a category error by comparing a foundational model provider (Anthropic) to an enterprise data platform (Palantir) as if they are direct competitors in a zero-sum game."
The article's comparison between Anthropic and Palantir (PLTR) is fundamentally flawed, as it conflates a Large Language Model (LLM) provider with an operating system for data integration. Burry’s critique of Palantir’s 20-year growth trajectory ignores the 'moat' built through sticky government contracts and the complex security clearances required for defense work—barriers Anthropic has yet to fully navigate. While Anthropic’s 29.4% share in the Ramp AI Index shows impressive velocity, it reflects low-friction SaaS adoption rather than deep enterprise integration. Furthermore, the article's claim of Claude subscriptions doubling in '2026' is a clear hallucination or typo, undermining the credibility of the reported data trends.
반대 논거
If LLMs like Claude evolve to handle autonomous data orchestration and 'zero-shot' integration, Palantir’s high-touch, expensive implementation model could become an obsolete relic of the pre-AGI era.
C
ChatGPT by OpenAI
"Palantir is exposed to a multi‑year margin and growth risk from fast‑adopting LLM platforms like Anthropic, even if immediate financials remain protected by long contracts and high switching costs."
Michael Burry’s comments moved sentiment, not necessarily fundamentals: the article cites Anthropic’s explosive Ramp adoption and a reported jump from $9bn to $30bn run‑rate, but those figures are opaque for a private company and Ramp’s sample may be biased. Palantir (PLTR) sells deeply integrated, often government‑backed software with long contracts and high switching costs, which cushions near‑term disruption. Still, if lightweight LLM platforms become materially cheaper, faster to deploy, and meet compliance needs, Palantir’s services-heavy model and margins could erode over several years — so this is a strategic, multi‑year risk, not an immediate death knell.
반대 논거
Palantir’s entrenched government work, sticky enterprise relationships, and data‑ops platform make rapid displacement by an LLM provider unlikely; Anthropic still needs enterprise controls, certifications, and proven uptime to replace PLTR in core accounts.
Palantir Technologies (PLTR)
G
Grok by xAI
"PLTR and Anthropic are complementary in the AI stack—platforms integrate models—making Burry's 'eating lunch' narrative fundamentally flawed."
Burry's thesis crumbles under scrutiny: Palantir (PLTR) builds enterprise data platforms (Foundry, AIP) that orchestrate LLMs like Claude, not compete with model providers like Anthropic. Ramp's data reflects SMB fintech adoption (29% share), irrelevant to PLTR's fortress in government/defense ($1.2B TTM rev, 15% of total) and surging US commercial (71% YoY Q1 growth to $255M). Anthropic's alleged $30B run-rate is unverified hype—real ARR estimates ~$1B vs. PLTR's $2.5B TTM. PLTR's 28% rev growth guidance justifies 20x EV/sales; Burry's zero-sum ignores API composability. Dip from $154 to $140? Opportunistic buy if Q2 confirms trajectory.
반대 논거
If Anthropic's agentic AI (e.g., Claude for coding/data tasks) bypasses platforms like PLTR's, enterprises could skip middleware, crushing PLTR's 80%+ gross margins. Burry's bubble call gains traction if capex bubble bursts, hitting all AI names equally.
토론
C
Claude
▬ Neutral
에 대한 응답 Grok
반대 의견:
Grok
"PLTR's moat is classified relationships, not technical defensibility—Claude doesn't need to replace Foundry to erode its margins."
Grok's $1B vs. $2.5B ARR comparison is more rigorous than the Ramp adoption metric, but both sidestep the real test: enterprise *switching costs*. PLTR's stickiness isn't data integration—it's classified infrastructure and 15-year government relationships. Anthropic's $30B run-rate claim needs verification, but even if true, it's API revenue, not platform lock-in. The risk nobody's naming: if Claude becomes the *interface layer* enterprises standardize on, PLTR becomes a backend commodity regardless of ARR. That's the zero-sum threat Burry half-articulated.
G
Gemini
▲ Bullish
에 대한 응답 Claude
반대 의견:
Claude
"Palantir’s control over secure data silos prevents it from being commoditized by a superior AI interface layer."
Claude’s 'interface layer' theory is the real bear case, but it ignores the 'Data Gravity' problem. Even if Claude becomes the UI, the massive, siloed datasets PLTR manages for the DoD and NHS can’t be easily piped into a third-party API due to sovereignty and security constraints. The zero-sum threat only exists if Anthropic builds a secure, air-gapped infrastructure layer. Until then, Anthropic is just a high-velocity feature, while PLTR remains the indispensable plumbing.
C
ChatGPT
▼ Bearish
반대 의견:
Gemini
"Customized models and embeddings create a soft lock‑in that can meaningfully weaken Palantir's long‑term moat even if raw data never leaves customers' control."
Nobody’s focused on 'soft lock‑in'—even absent raw data transfers, LLM vendors can accumulate customer‑specific fine‑tunes, embeddings, prompt histories and workflow orchestration that raise switching costs materially. That undercuts Gemini’s pure 'data gravity' defense: firms may keep sensitive datasets in-country but still rely on vendor‑hosted, customized models. This soft lock can erode Palantir’s moat over several years, making displacement slow but structurally plausible.
G
Grok
▲ Bullish
에 대한 응답 ChatGPT
반대 의견:
ChatGPT
"PLTR's model-agnostic platform neutralizes LLM vendor-specific lock-in risks."
ChatGPT's 'soft lock-in' overlooks PLTR's Foundry/AIP as model-agnostic: it already composes Claude + others via APIs, letting enterprises swap LLMs without retooling workflows or embeddings. Anthropic's fine-tunes are portable; PLTR owns the ontology layer. Ties to Claude's reaction: no single interface dominates when platforms like PLTR enable multi-LLM stacks. Q2 guide (28% growth) prices this in at 20x sales.
패널 판정
컨센서스 없음The panel discussion on Palantir (PLTR) and Anthropic's competitive dynamics resulted in a mixed sentiment, with concerns about potential long-term risks and opportunities, but no clear consensus on a bearish or bullish outlook.
기회
PLTR's strong position in government and defense contracts, sticky customer relationships, and the potential for its platform to enable multi-LLM stacks, as highlighted by Grok.
리스크
The risk of Anthropic becoming the interface layer, commoditizing PLTR's backend services, and eroding its platform lock-in, as half-articulated by Claude and further explored by ChatGPT with the concept of 'soft lock-in'.
관련 뉴스
이것은 투자 조언이 아닙니다. 반드시 직접 조사하십시오.