The Artificial intelligence (AI) business is moving fast, and simply building a chatbot that answers questions is no longer a golden ticket. Tech companies are now racing to create AI assistants that can actually get things done, and the competition is heating up. Meta (META) has Muse, OpenAI has Dots, and now Anthropic, the AI research company, wants a seat at the same table.
On Wednesday, Anthropic unveiled Claude Duet, its latest AI agent designed to help users manage workplace tasks with less hand-holding. Built on its Opus 5.5 model, Duet aims to move beyond answering prompts and become a more proactive assistant that can work across connected applications. The goal is to spend less time buried in emails, spreadsheets, and endless to-do lists, and more time getting meaningful work done.
But this launch arrives at an interesting moment, with Anthropic reportedly preparing for a potential IPO in November. As investors watch the AI race intensify, can Claude Duet help the company stand out from its increasingly formidable rivals?
About Anthropic
San Francisco-based Anthropic has gone from a relatively young AI startup to one of the biggest names in the business in just a few years. Founded in 2021, the company is best known for Claude, its family of AI chatbots and models, and Claude Code, a tool that helps developers write software. While everyday users pay for Claude subscriptions, businesses, especially technology companies, bring in most of the money.
That demand has helped Anthropic reportedly reach a whopping $67 billion annualized revenue run rate in 2026, less than six years after its launch. Not bad for a company that has not even been around for a decade. Beyond making AI useful for work, Anthropic also focuses on making it safer and more reliable. As a public benefit corporation, it aims to build powerful AI while managing its risks, and its rapid growth has made it a major name to watch in the AI boom.
About Claude Duet—Anthropic Wants to Be Your Coworker, Not Just Your Chatbot
Think of regular Claude as the colleague you turn to when you need an answer, help writing an email, or a report put together. Claude Duet, meanwhile, is the colleague who keeps working even after you have called it a day. See the difference? Anthropic wants to move beyond the usual chatbot routine and build an AI assistant that actually takes work off your plate.
Powered by Anthropic’s Opus 5.5 model, Duet is a proactive agent that uses the existing Claude account’s saved context, preferences, and connected applications to prepare morning briefings, flag important developments, schedule meetings, create financial forecasts, and even build dashboards and web apps. It runs inside its own virtual machine with a browser, so assigned tasks can continue while your phone or laptop is switched off. Finally, an assistant that does not clock out just because you do.
With access to over 6,000 connectors and plugins, including Gmail, Slack, QuickBooks, Stripe, and Shopify, Duet aims to make managing multiple workplace tools considerably less exhausting. It also seeks approval before taking consequential actions, such as sending messages or making payments.
Currently, Claude Duet is available through a waitlist to eligible U.S. subscribers on Anthropic’s Pro and Max plans, with no separate charge announced during the research preview.
Anthropic is reportedly preparing for a potential IPO, and expanding beyond conventional chatbots could strengthen its growth story. If Duet can convince businesses that AI can do more than generate impressive answers, Anthropic might have another compelling selling point for investors. Because in the race to make AI indispensable, being smart is great, but getting actual work done is what counts.
How Does It Stack up Against the Competition?
Meta’s Muse is primarily aimed at everyday consumers, helping them manage personal tasks, while OpenAI’s Dots also focuses on ongoing projects, connected apps, and learning user preferences. Duet’s distinction is its emphasis on workplace productivity, particularly for professionals, entrepreneurs, and small-business owners. Of course, it is not perfect. The assistant is still experimental, checks connected accounts every few hours rather than continuously, and can occasionally misjudge what deserves attention.
Anthropic’s Financials: Explosive Revenue but Massive Losses
With Claude Duet now entering the AI agent race, Anthropic is giving investors another reason to pay attention. But beyond the latest product launch and the battle against Meta and OpenAI, there’s a bigger story brewing—the company’s financials. With a potential IPO on the horizon, it is time to look beyond the AI excitement and see what the numbers actually reveal. And, well, there is plenty to unpack.
Anthropic’s reported draft S-1 paints a picture of a business growing at an extraordinary pace. Revenue reportedly reached $16.3 billion in the first half of 2026—that's almost 15x from just $1.1 billion a year earlier. It yields an annual run rate of about $67.2 billion. The momentum continued into the third quarter, when revenue climbed 45% sequentially to $16.8 billion. Management also reportedly told prospective investors that the company achieved adjusted profitability during the quarter. That’s impressive, absolutely. But there’s more to this story than revenue racing higher.
Running advanced AI models is not cheap, and Anthropic’s infrastructure spending is growing alongside its business. Computing and infrastructure costs reportedly approached $9 billion in the first half of 2026, compared with $2.3 billion a year ago. The draft also outlines approximately $450 billion in mostly non-cancelable cloud agreements spanning seven to 10 years. Adding in chip commitments, infrastructure obligations reportedly exceed $600 billion. Anthropic also expects to spend another $750 billion to secure additional computing capacity, including deals still under negotiation.
Then comes the headline-grabbing loss—$80.7 billion in the first half. However, approximately $72 billion reportedly came from non-cash financing-related adjustments, so that figure should not automatically be mistaken for cash burn. Operating losses, cash flow, and financing terms deserve a closer look.
Customer concentration presents another challenge. Anthropic’s two largest customers reportedly accounted for around 20% of first-half revenue, with significant dependence on a limited customer base expected to persist.
Meanwhile, investors have reportedly discussed a $1.5 trillion valuation, below an earlier $2 trillion target. Neither figure represents finalized IPO pricing.
Final Thoughts on Anthropic
Claude Duet gives Anthropic another way to stand out as Meta’s Muse, and OpenAI’s Dots compete for a place in the AI agent race. But with a potential IPO on the horizon, the company has more to prove than just launching another promising product. Skyrocketing revenue is encouraging, but massive computing commitments, customer concentration risks, and questions around sustainable profitability complicate the picture. Anthropic has the growth, the technology, and the investor attention. Now, it needs to show that it can turn all that momentum into lasting profits.
On the date of publication, Sristi Suman Jayaswal did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. For more information please view the Barchart Disclosure Policy here.