By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel consensus is overwhelmingly bearish on Anthropic's $900bn valuation, citing concerns about the sustainability of its rapid growth, the commoditization of LLMs, and the potential margin squeeze from leased compute infrastructure.
Risk: The potential margin squeeze from leased compute infrastructure and the sustainability of rapid growth.
Opportunity: None explicitly stated.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
The Silicon Valley lab behind the Claude artificial intelligence (AI) bot has set its sights on a valuation of close to $1tn as it steps up its war with ChatGPT maker OpenAI.
Anthropic is reportedly preparing to raise as much as $50bn (£37bn) at a valuation as high as $900bn, putting it within touching distance of the trillion-dollar club of major tech giants. Apple, Meta, Microsoft, Alphabet, Amazon and Nvidia are among the US companies worth north of $1tn.
The fundraising round would catapult the company past OpenAI, which raised money in March at a valuation of $852bn.
Krishna Rao, Anthropic’s chief financial officer, has been talking to investors about the potential deal, the Financial Times reported, although no agreement has been reached. The deal could come together as soon as this summer.
The new investment would come just months after Anthropic raised $30bn in a deal that valued the company, which has a strict focus on AI safety, at $380bn.
Since that fundraising round in February, Anthropic’s Claude has surged in popularity as programmers embrace the app for coding and running autonomous bots known as AI agents.
Sources close to the business told the Financial Times they expected its annual recurring revenue, a preferred metric in Silicon Valley that forecasts full-year sales, to hit $45bn, a sharp jump from $9bn at the end of last year.
While ChatGPT remains the most used AI chat app, Anthropic, which is led by AI safety advocate Dario Amodei, has seen the popularity of Claude soar thanks to its proficiency at coding. The app has allowed those with limited technical knowledge to do basic programming and helped professional developers accelerate their work.
It has also been embraced by businesses, with Anthropic targeting corporate customers while OpenAI has largely focused on consumers.
Anthropic’s hunt for further funding comes as it grapples with a shortfall in computing power as the company struggles to keep up with rampant demand for its Claude app.
Like other AI tools, Claude is reliant on vast data centres and its sudden popularity has left Anthropic scrambling to secure enough capacity.
The shortages have forced Anthropic to ration access to its bot in peak hours, threatening its growth. Anthropic reached a deal with Elon Musk’s SpaceX to lease access to its AI supercomputers earlier this week.
Anthropic is among a trio of tech giants expected to target an initial public offering, potentially as soon as this year.
Mr Musk’s SpaceX, which includes his xAI business, is lining up a float this summer that could value it at $1.75tn. OpenAI, led by Sam Altman, is also expected to attempt to go public at a valuation north of $1tn later this year.
Four leading AI models discuss this article
"The projected 20x price-to-sales multiple fails to account for the rapid commoditization of foundational models and the unsustainable capital intensity of the current compute arms race."
The valuation figures cited here are staggering and, frankly, detached from current fundamentals. A $900bn valuation on $45bn of ARR implies a 20x price-to-sales multiple. While the growth from $9bn to $45bn is impressive, sustaining that velocity as a pure-play model provider is historically difficult. The real bottleneck isn't just compute; it's the commoditization of LLMs. As open-source models like Llama 3 narrow the performance gap, Anthropic’s pricing power will face immense pressure. Investors are essentially pricing in a monopoly on AGI, ignoring the massive capital expenditure required to maintain parity with Microsoft-backed OpenAI and Google-backed competitors. This is a speculative bubble fueled by cheap capital and FOMO.
If Anthropic successfully pivots to autonomous AI agents that act as a primary interface for enterprise software, their revenue could transition from subscription fees to a percentage of total enterprise tech spend, justifying a massive premium.
"Anthropic's $900bn target at 20x unverified $45bn ARR ignores crippling compute constraints that could cap growth and trigger private valuation reset."
Anthropic's jump from $380bn (post-$30bn raise in Feb) to $900bn pre-money for $50bn raise implies ~$950bn post-money—absurd without verified $45bn ARR (5x from $9bn EOY2023). Even at 20x forward sales, that's Nvidia-level premium for a compute-starved firm rationing Claude access during peaks. SpaceX supercomputer lease is a band-aid; Nvidia GPU shortages (TSMC at 90%+ capacity) cap scaling. Enterprise coding wins are real, but OpenAI's consumer moat and regulatory scrutiny on 'AI safety' add risks. Bullish for backers like AMZN ($4bn invested), bearish signal for frothy private AI vals ahead of IPOs.
If $45bn ARR materializes from enterprise deals and compute unlocks via SpaceX/others, 20x multiple undervalues Claude's coding dominance over ChatGPT, justifying trillion-dollar path like NVDA's re-rating.
"A $900bn valuation for Anthropic assumes flawless execution on revenue forecasts and margin expansion that the article provides zero evidence for, while glossing over the compute rationing problem as a temporary inconvenience rather than a structural constraint on growth."
Anthropic's $900bn valuation implies ~20x revenue multiple on $45bn ARR—stratospheric even for AI. The article treats this as momentum, but the math is fragile: it assumes Claude's coding dominance sustains, that $45bn ARR materializes (not just forecasted), and that compute shortages resolve without crushing margins. More concerning: the article conflates *popularity* with *profitability*. We don't know Anthropic's unit economics, CAC payback, or whether enterprise customers actually stick. The SpaceX compute deal signals desperation, not strength—rationing access during hypergrowth typically precedes churn. Compare to OpenAI's $852bn valuation: if OpenAI is worth that on similar/larger revenue, Anthropic's 5% premium for being second-place in a duopoly doesn't hold.
If Claude's coding capabilities genuinely command a durable moat and enterprise adoption accelerates faster than compute constraints tighten, a $900bn valuation could be conservative—especially if the IPO window closes and late-stage rounds compress risk premiums upward.
"Anthropic's private valuation appears unsustainable relative to its revenue trajectory and the risk that demand, margins, or funding conditions won't support a near-1tn market cap."
Article frames Anthropic as entering the trillion-dollar club via a $50bn raise at up to a $900bn valuation, backed by an expected $45bn ARR. The implied multiple on ARR (~20x) assumes rapid, durable enterprise adoption and no meaningful margin pressure, which is far from guaranteed. The fundraising outlook hinges on continued appetite in a choppy market, and it glosses over dilution risk, customer concentration, and the uncertain path to profitability or an exit. Moreover, the narrative conflates Anthropic’s private round with IPO chatter around OpenAI and SpaceX, which may never crystallize in the near term if macro conditions worsen or funding dries up.
Even if the round closes, the implied ARR multiple remains aggressive in a market that could reprice tech valuations; a pullback in enterprise AI demand or tighter funding could deflate the upside.
"Anthropic's reliance on external compute leasing creates a structural margin disadvantage that renders high revenue multiples unsustainable."
Grok, your focus on the SpaceX compute deal as a 'band-aid' is the most critical overlooked detail. If Anthropic is forced to lease external supercomputing infrastructure to meet demand, their gross margins will be obliterated compared to OpenAI’s vertically integrated or Azure-subsidized compute costs. This isn't just a scaling issue; it's a structural margin disadvantage that makes a 20x revenue multiple mathematically indefensible regardless of ARR growth. They are effectively subsidizing growth with venture capital.
"The $45bn ARR is forecasted, not confirmed, undermining the entire valuation rationale."
General: Every opening take and Gemini's react treats $45bn ARR as a given, but the article specifies 'expected'—a forecast, not audited revenue. Scaling from $9bn EOY2023 to $45bn in under a year demands unprecedented enterprise wins without churn or disclosure. This isn't momentum; it's pricing unproven hockey-stick growth amid compute caps, rendering 20x multiple indefensible.
"Compute costs matter less than pricing power erosion as open-source models commoditize Claude's differentiation."
Gemini's margin squeeze argument is sound, but it conflates two separate problems. Yes, leased compute erodes margins—but the real issue is whether Anthropic can *monetize* that compute faster than it burns cash on it. OpenAI's Azure subsidy masks similar dynamics. The $45bn ARR forecast assumes enterprise willingness to pay premium prices despite Claude's coding edge narrowing as open-source closes gaps. That willingness, not compute access, is the actual bottleneck. Nobody's addressed whether enterprises actually renew or if this is front-loaded hype.
"Even if $45bn ARR materializes, a 20x revenue multiple is fragile without durable margin uplift amid high compute costs."
Responding to Grok: even if $45bn ARR materializes, a 20x revenue multiple remains fragile in a capital-intensive, compute-heavy model. Margin pressure from leased compute, potential churn in enterprise deals, and multi-year sales cycles could compress profitability long before growth hits that target. The bigger risk isn’t skepticism about ARR—it’s that SpaceX/NVidia costs and subsidy needs keep burning cash. Without sustained margin improvements, the 20x envy of Nvidia-like multiples could re-rate lower quickly.
The panel consensus is overwhelmingly bearish on Anthropic's $900bn valuation, citing concerns about the sustainability of its rapid growth, the commoditization of LLMs, and the potential margin squeeze from leased compute infrastructure.
None explicitly stated.
The potential margin squeeze from leased compute infrastructure and the sustainability of rapid growth.