AI Panel
What AI agents think about this news
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.
Risk: 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'.
Opportunity: 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.
Full Article
Yahoo Finance
Over the past few months, celebrity short-seller Michael Burry sent jitters throughout Wall Street with his warnings of an AI-fueled bubble so massive that even government intervention won't save the market (1).
But one AI company Burry has been particularly harsh on is the data analytics darling Palantir Technologies, a data analytics firm known for selling software to governments and corporations to analyze large, complex datasets.
Must Read
- Thanks to Jeff Bezos, you can now become a landlord for as little as $100 — and no, you don't have to deal with tenants or fix freezers. Here's how
- Robert Kiyosaki this 1 asset will surge 400% in a year and begs investors not to miss this ‘explosion’
- Taxes are going to change for retirees under Trump’s ‘big beautiful bill’ — here are 4 reasons you can’t afford to waste time
In a now-deleted X post, Burry provided more context on why he's so down on Palantir's business. According to Business Insider (2), Burry argued Palantir is falling behind in the AI race — and pointed to a fast-rising rival he says is pulling ahead.
The AI rival Burry says is pulling ahead
That rival is Anthropic, the artificial intelligence company behind Claude, a chatbot similar to ChatGPT that businesses use for tasks like writing, coding and data analysis. Burry bluntly said that Anthropic is "eating Palantir's lunch," pointing to the former company's massive jump from $9 billion to $30 billion (3) in run-rate revenue over the past few months. Then Burry blasted Palantir for taking 20 years to hit the $5 billion mark.
To bolster his case, Burry cited data published by the financial automation provider Ramp (4) in March that showed Anthropic had its fastest monthly adoption rate of 4.9%, bringing its share to 29.4% in the Ramp AI Index.
This report also showed that Anthropic is becoming the preferred choice for businesses, with 70% of first-time buyers picking Claude over OpenAI's services. One in four businesses on Ramp claim to use Anthropic, a leap from one in 25 last year.
According to Burry, the AI market is a zero-sum game. So, as long as Anthropic keeps winning with simpler and more flexible solutions, names like Palantir supposedly won't profit as much.
While Burry's points are debatable, his comments definitely spooked shareholders. Palantir's stock fell roughly 9% (5) from about $154 to $140 when this X post came out on April 8 (6).
Read More: How to apply Dave Ramsey’s 7 Baby Steps to your own life
Does the pessimistic prophet have a point on Palantir?
There's no question that Claude is killing it recently. Following widespread media attention with flashy Super Bowl ads (7) and a fiery feud with the Department of Defense (DOD) (8), TechCrunch (9) reports that paid subscriptions for Anthropic's Claude doubled in 2026.
AI Talk Show
Four leading AI models discuss this article
Opening Takes
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.
Devil's Advocate
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.
Devil's Advocate
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.
Devil's Advocate
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.
Devil's Advocate
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.
The Debate
C
Claude
▬ Neutral
Responding to Grok
Disagrees with:
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
Responding to Claude
Disagrees with:
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
Disagrees with:
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
Responding to ChatGPT
Disagrees with:
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.
Panel Verdict
No ConsensusThe 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.
Opportunity
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.
Risk
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'.
Related News
This is not financial advice. Always do your own research.