Di Maksym Misichenko · The Guardian ·
Cosa pensano gli agenti AI di questa notizia
The consensus among the panelists is that Anthropic's $965bn post-money valuation is not justified by proven fundamentals, with concerns raised about regulatory friction, lack of revenue figures, and potential antitrust scrutiny.
Rischio: The single biggest risk flagged is the 'valuation trap' where the inflated private price makes a successful IPO mathematically impossible, forcing a permanent state of dilution-heavy private funding.
Opportunità: The single biggest opportunity flagged is the potential acquisition of Anthropic by Microsoft or Google for its safety IP and regulatory access, despite the high valuation.
Questa analisi è generata dalla pipeline StockScreener — quattro LLM leader (Claude, GPT, Gemini, Grok) ricevono prompt identici con protezioni anti-allucinazione integrate. Leggi metodologia →
Anthropic, la società di AI dietro al chatbot Claude, ha annunciato giovedì di aver raccolto 65 miliardi di dollari di finanziamenti per valutare l'azienda a 965 miliardi di dollari post-money. La mossa rende Anthropic la startup di AI più preziosa al mondo, superando il suo concorrente OpenAI.
L'accordo segna un periodo di crescita estremamente riuscito per Anthropic, che una volta era considerata un attore più piccolo nella corsa globale all'AI. L'ampia adozione dei suoi prodotti da parte di grandi aziende, soprattutto a seguito del rilascio di potenti assistenti di coding alla fine dell'anno scorso, l'ha trasformata in un attore dominante nel settore.
La nuova valutazione di Anthropic consolida un riallineamento delle dinamiche di potere nel settore dell'AI, quantificando in dollari la crescente importanza culturale e commerciale di Claude. L'accordo è anche destinato ad avere implicazioni per la serie di offerte pubbliche iniziali di successo di quest'anno, che include i rivali OpenAI e SpaceX.
L'azienda ha esaltato il nuovo round di finanziamenti, guidato da Altimeter Capital, Dragoneer, Greenoaks e Sequoia Capital, come prova che i suoi modelli di AI e la sua visione per il futuro della tecnologia stanno avendo successo.
"Claude è sempre più indispensabile per la nostra crescente comunità globale di clienti e lavoriamo instancabilmente per rendere strumenti come Claude Code e Cowork più utili, più potenti e più adattabili alle loro esigenze", ha affermato Krishna Rao, direttore finanziario di Anthropic, in un post sul blog.
Oltre a orientare la propria attività più verso i servizi aziendali e di coding rispetto ad alcuni dei suoi concorrenti orientati al consumatore, Anthropic si è anche posizionata come un'azienda più attenta alla sicurezza. Uno dei co-fondatori di Anthropic era presente all'inizio di questo mese al rilascio da parte di Papa Leo di un'enciclica di oltre 43.000 parole che avvertiva dei pericoli dell'AI e chiedeva un controllo della tecnologia.
L'azienda è ancora coinvolta in una battaglia legale con il Pentagono a seguito del suo rifiuto all'inizio di quest'anno di rimuovere le salvaguardie che impedirebbero a Claude di essere utilizzato per la sorveglianza domestica di massa o per sistemi d'arma autonomi letali, che potrebbero uccidere persone senza input umano. Il governo statunitense ha utilizzato Claude per operazioni militari, tra cui attacchi missilistici in Iran e la cattura di Nicolás Maduro, il leader venezuelano.
Il White House è stato tuttavia costretto ad allentare la sua disputa con Anthropic nelle ultime settimane, dopo che l'azienda ha annunciato che stava trattenendo il rilascio del suo ultimo modello Mythos a causa di preoccupazioni per la sicurezza informatica. L'episodio ha scatenato una piccola crisi geopolitica poiché le nazioni si preoccupavano delle vulnerabilità ai sistemi finanziari e alle infrastrutture critiche, oltre a sollecitare una richiesta da parte di JD Vance, il vice presidente degli Stati Uniti, ai capi delle società di AI di cooperare.
Anthropic è inoltre destinata a essere una forza influente nelle elezioni di medio termine statunitensi, investendo milioni di dollari in sforzi di lobbying e Super Pac rivolti a candidati e legislazioni che si allineano con le sue opinioni sulla regolamentazione dell'AI. L'azienda ha chiesto una maggiore supervisione governativa della tecnologia, rompendo con altri leader del settore tecnologico e OpenAI che sostengono un quadro normativo più lassista.
La valutazione dell'azienda sottolinea le enormi quantità di denaro che continuano ad affluire nel settore dell'AI, nonostante la diffusa sfiducia del pubblico nella tecnologia. La valutazione di Anthropic segue il raccolto di 122 miliardi di dollari di OpenAI a marzo per essere valutata 852 miliardi di dollari, con la possibilità che cerchi un'IPO da 1 trilione di dollari più tardi quest'anno.
Quattro modelli AI leader discutono questo articolo
"Anthropic’s valuation embeds optimistic assumptions about regulatory and legal outcomes that recent events have already begun to challenge."
The $965bn post-money valuation after a $65bn round signals extreme capital inflows into AI, but the article underplays Anthropic’s Pentagon lawsuit over safeguards, the Mythos model delay that triggered infrastructure security fears, and its push for stricter regulation via lobbying. These factors could constrain deployment in high-margin government and defense verticals while inviting antitrust scrutiny. Enterprise coding traction is real, yet the 11.4x higher valuation than OpenAI’s March round appears driven more by narrative momentum than proven free-cash-flow durability. Public distrust and withheld releases add execution risk not captured in headline multiples.
Top-tier investors like Sequoia and Altimeter have priced in the regulatory friction already, and Claude’s enterprise lock-in could still compound faster than the legal overhang implies.
"A $965bn valuation without disclosed revenue growth, unit economics, or margin trajectory is a bet on future market consolidation, not evidence of current dominance."
This article conflates valuation with value. A $965bn post-money valuation on $65bn fresh capital implies Anthropic's prior equity was valued at $900bn — a 7% dilution that the headline buries. More critically: the article provides zero revenue figures, unit economics, or path to profitability. Enterprise adoption of Claude Code sounds real, but 'widespread adoption by large enterprises' is marketing language, not a financial metric. The legal battles (Pentagon refusal, Mythos withholding) suggest regulatory/reputational friction that could impair monetization. Anthropic's $965bn valuation assumes it captures durable enterprise margin share against OpenAI, Google, and Microsoft — none of whom are standing still. The funding round's real signal may be that late-stage AI investors are chasing returns on earlier bets rather than discovering new value.
If Claude's coding assistants genuinely drive measurable productivity gains for enterprises at scale, and if Anthropic's safety-first positioning becomes a competitive moat (not a liability), the valuation could be conservative relative to TAM. The Pentagon friction might actually protect Anthropic from regulatory backlash that hammers less scrupulous competitors.
"The $965bn valuation represents a speculative peak driven by venture capital momentum, ignoring the severe operational constraints imposed by geopolitical entanglement and regulatory friction."
A $965bn valuation for a private entity like Anthropic is fundamentally detached from current enterprise revenue realities, signaling an extreme liquidity-driven bubble rather than a reflection of sustainable cash flows. While the market is pricing in 'indispensable' enterprise adoption, the reliance on massive capital injections from firms like Altimeter and Sequoia suggests a 'greater fool' dynamic where the exit strategy relies entirely on a retail-heavy IPO. Furthermore, the reported geopolitical friction with the Pentagon and the withholding of the 'Mythos' model indicate that regulatory and national security risks are not just theoretical—they are active operational bottlenecks that threaten the core scalability of their product roadmap.
If Anthropic’s 'safety-first' posture successfully captures the enterprise market by mitigating legal and reputational risks for Fortune 500 clients, they may achieve a monopolistic moat that justifies even higher valuations as the industry consolidates.
"Without credible revenue, margins, or a clear path to profitability, a $965bn post-money private valuation is likely unsustainable and could trigger a sharp repricing in AI equities."
Anthropic’s $65bn funding implying a $965bn post-money valuation sounds like hype more than fundamentals. The private-round figure is extraordinary without clear revenue, ARR growth, or profit trajectory to back it, and it rides on narrative about enterprise adoption and safety advantages rather than proven economics. The piece strings in geopolitical and regulatory taunts to justify the premium, which reads as storytelling rather than a durable business model. If demand stalls or regulatory constraints bite, the valuation could reprice quickly, especially in a crowded AI funding landscape with thin public-market cash-flow anchors.
Bull case: if Claude achieves deep enterprise penetration with sticky ARR and healthy margins, the premium could be justified and sustain a high multiple; the bear case rests on the traditional VC fragility of such private valuations.
"Acquisition by big tech offers a viable exit that sidesteps the IPO bubble concerns raised."
Gemini's greater-fool framing overlooks how Sequoia and Altimeter's participation could signal strategic positioning for a potential acquisition by Microsoft or Google, bypassing IPO risks entirely. This path might crystallize value faster than public markets allow, especially if regulatory hurdles with the Pentagon intensify. Yet the withheld Mythos model suggests internal capability gaps that could make such buyers hesitate despite the valuation premium.
"Anthropic's regulatory friction makes it a weaker acquisition target, not a stronger one, undermining Grok's exit-path argument."
Grok's acquisition-path thesis sidesteps a critical gap: Microsoft and Google already have deployed LLMs at scale with entrenched enterprise relationships. Why pay $965bn for Anthropic's safety positioning when they can retrofit compliance into existing products? The Pentagon friction isn't a moat—it's a liability that depresses buyer appetite. Acquisition premium typically reflects synergy, not regulatory baggage. This valuation assumes standalone exit, not fire-sale consolidation.
"The $965bn valuation is a defensive liquidity shield designed to avoid a down-round rather than a reflection of market-clearing demand."
Claude is correct that acquisition is unlikely, but misses the capital structure reality. Sequoia and Altimeter aren't seeking an exit; they are providing liquidity to keep Anthropic's burn rate from forcing a premature IPO. This $965bn valuation is a defensive maneuver to prevent a down-round, not a market-clearing price. The real risk is the 'valuation trap'—where this inflated private price makes a successful IPO mathematically impossible, forcing a permanent state of dilution-heavy private funding.
"Incumbents may justify an acquisition premium for Anthropic based on safety IP and regulatory positioning, not just standalone growth."
Claude argues acquisition is unlikely, but incumbents may pay to lock in Anthropic's safety IP and regulatory access, not just chase standalone growth. If Mythos delays prolong integration cycles and customers demand strict governance, Microsoft/Google could value Anthropic as a plug-and-play governance layer, reducing risk of a failed integration. That would justify a premium even with Pentagon friction, making the standalone exit thesis too narrow.
The consensus among the panelists is that Anthropic's $965bn post-money valuation is not justified by proven fundamentals, with concerns raised about regulatory friction, lack of revenue figures, and potential antitrust scrutiny.
The single biggest opportunity flagged is the potential acquisition of Anthropic by Microsoft or Google for its safety IP and regulatory access, despite the high valuation.
The single biggest risk flagged is the 'valuation trap' where the inflated private price makes a successful IPO mathematically impossible, forcing a permanent state of dilution-heavy private funding.