Model

Anthropic falls 11% toward its $83 IPO price as Microsoft undercuts Claude

Anthropic fell about 11% Friday morning, back toward the $83 IPO price, as Microsoft's price attack on Claude met the open market. Microsoft holds Anthropic's compute contract, its shares and a cheaper rival product; a leveraged income ETF on the two-day-old stock listed the same morning.

Richard TangRichard Tangthree-minute read
Share
Microsoft chief executive Satya Nadella, in glasses and a dark jacket, photographed against a white background
Satya Nadella, Microsoft's chief executive, whose cheaper Copilot pricing opened two days before Anthropic's Nasdaq listing and met the open market on its second session.

Anthropic's$350B — Anthropic, private, latest valuation $350B second session as a public company is unwinding its debut. The stock was indicated about 11% lower on Friday morning, November 20 1, sliding back toward the $83 a share it sold at in the largest IPO on record 2 3. The new pressure is a price attack Microsoft−1.86% — Microsoft, down 1.86 percent today opened two days before the listing, and Microsoft has arranged to be paid whichever way the fight goes. That, not the daily tick, is the story: cheaper AI seats help every buyer, while the risk has been left with the newest shareholders.

The cheapness is a router, not a benchmark crown

Copilot's undercut is mechanical. It routes each request to the cheapest model that can do the job, Microsoft's own MAI models included, rather than parking every user on one costly frontier model 4 5. At Microsoft's Ignite conference on November 17, Judson Althoff, who runs its commercial business, said anything you can do in Claude or ChatGPT "you can do better with Copilot, and for less" 5.

The evidence is a GigaOm study Microsoft commissioned: everyday business use of Claude at $205 a month, against $30 for Copilot 5. Anthropic counters that its non-coding enterprise users spend under $75 a month 1. The two figures do not measure the same thing, but the direction of enterprise pricing points only one way.

Microsoft is selling Copilot at $30 a seat against $205 for everyday Claude use

$0/mo$50/mo$100/mo$150/mo$200/mo$250/moCopilot (GigaOm study)$30/moClaude, non-coding enterprise (Anthropic)$75/moClaude, everyday business use (GigaOm)$205/mo
Data
Value
Copilot (GigaOm study)$30/mo
Claude, non-coding enterprise (Anthropic)$75/mo
Claude, everyday business use (GigaOm)$205/mo
Monthly price per user, US dollars, as each side describes it. The $30 Copilot figure and the $205 Claude figure come from a GigaOm study Microsoft commissioned; the $75 is Anthropic's own figure for non-coding enterprise users. The populations differ, so the bars compare the pitches, not like-for-like bills. Sources: GeekWire on the GigaOm study; Yahoo Finance.5,1

Microsoft holds both sides of the fight

Copilot is fighting Claude for corporate seats while Microsoft stays Anthropic's landlord and shareholder. Anthropic has committed to $30 billion of Azure compute, and Microsoft agreed to invest up to $5 billion in the company 1.

The stake handed Microsoft a $3.2 billion gain in its June quarter, and Microsoft is also among seven buyers in a $26 billion private placement at the $83 IPO price 1 2. Copilot sells Claude itself: "you can route across all of that," chief executive Satya Nadella told Jim Cramer, adding, "We're building tools, not gods" 1.

The margin scorecard behind that pitch is guided to hold within a point this fiscal year 1.

Microsoft's operating margin climbed through the AI buildout to 46.8%

Margin hit 46.8% in fiscal 2026; fiscal 2027 guided to fall less than a point

40%42%44%46%48%FY2021FY2022FY2023FY2024FY2025FY202646.8%
Data
Operating margin
FY202141.6%
FY202242.1%
FY202341.8%
FY202444.6%
FY202545.6%
FY202646.8%
Operating margin, percent of revenue, fiscal years ending June. Sources: Microsoft SEC filings.8

Income wrapped around a two-day-old stock

Purpose Investments, a Canadian asset manager, listed ANTY on Cboe Canada the same morning: one stock underneath, covered calls on about half the book, roughly a quarter of leverage, monthly distributions 6. It launched beside a Micron−11.70% — Micron, down 11.70 percent today twin as memory names slid about 5% 7. On a share that can fall 11% by breakfast 1, the income is option premium, and the leverage deepens exactly the mornings it hurts most. The fee line is Purpose's. The drawdowns are the buyer's.

The line to watch is $83

Hold the IPO price and the anchor book sits near break-even; lose it and every buyer since the open is under water. The cheapness is the point: routing each task to the model that costs least is how AI reaches every office, and Microsoft has taxed both ends of that chain. The premium on a two-day-old ticker was never the technology, and it is the part giving way.

Get Model by email, free

The AI labs, their models, and the business behind them. In our daily email, every weekday.

Loading MSFT

Reader comments

0 comments

    Model

    HeyGen sells Avatar 6 while Tavus's Griffin scores 3.83 against a human's 3.92 and still has no price

    HeyGen put Avatar 6 on general availability on November 17 with a public API and first place in New Measure's blind Avatar Arena; Tavus's Griffin, which Tavus says is the highest score ever recorded for an AI on NVIDIA's face-to-face benchmark, is still a preview with no price. Revenue accrues to HeyGen today because Avatar 6 is the model on sale; Tavus's best model is not.

    Model

    Pika's Enact lets creators direct AI video mid-scene, with answers in under a second

    Enact, posted November 18, lets a creator drag an AI actor mid-scene, talk to the cast by name and redo the camera after locking a performance. Pika open-sourced the model underneath and kept the directing layer closed, a bet that value in AI video is moving from model quality to workflow.

    Model

    Ramp's invoices show Anthropic held 58.7% of US business AI spend vs OpenAI's 37.2% before its record IPO

    Ramp's index of 85,000 US business bill-payers is the first independent check on the record $81.3 billion IPO: the enterprise lead over OpenAI is real, and so is the falling price of what Anthropic sells.

    Model

    Google's Gemini 4 Neon took the coding lead the day before Anthropic's $1.5 trillion debut

    Google's mid-tier model debuted at No. 1 on Artificial Analysis's coding-agent index at $6.86 a task against Claude Opus 5.5's $12.61, a day before Anthropic's Nasdaq debut. A $1.5 trillion valuation built on coding-led growth now meets a measured, cheaper rival, and the exposure sits with whoever paid $83.

    Model

    ElevenLabs, valued at $22 billion, puts its voice agents inside NICE, Genesys, Amazon Connect and Five9

    Salesforce and Five9 have put more than $3 billion into buying customer-service AI this month. ElevenLabs, valued at $22 billion after a September share sale, embedded its agents natively into Five9's platform and three rivals instead, and the evidence points to pricing power sitting with the channel, not the model maker.

    Model

    OpenAI's agent bill now runs at twice its researchers' cash pay

    OpenAI's own report prices agent use at twice its researchers' cash pay while internal deployment is 2 percent of research compute, and the number lands the week Anthropic lists and OpenAI settles for half its raise. The bill stacks inference on the chip budget, and holders pay it.

    Subscribe to TheInference for free