Anthropic, the US company behind Claude, is preparing to list on Nasdaq in what could be one of the biggest floats ever. So why would it matter to someone in Melbourne with a mortgage and a super fund? If Anthropic joins a global share index your fund tracks, you could end up owning a small piece of it. The float could also affect what investors are willing to pay for the chipmakers and cloud providers that some Australian funds already hold.
This article builds on the first episode of our new series, Financial Autonomy in Conversation, where Paul Benson and I argued the bull and bear cases for the float. Listen below. (Figures are as at 4 November 2026, before the listing, and may change.)
What is the Anthropic IPO?
Anthropic is preparing to list on the Nasdaq under the ticker ANT.
As at 4 November 2026, there's no confirmed date. Reports point to mid to late November, after the US midterm elections on 3 November. Paul's read is that the timing isn't accidental, as the company may prefer to list once the election uncertainty has passed.
Press reports suggest a raise of around US$100 billion at a valuation of US$1.5 to 2 trillion, which would mean selling roughly 5 to 7% of the company. Those are reported estimates. The final terms may differ.
What the prospectus does show is how fast the business has grown:
Revenue of US$4.6 billion for the whole of 2025
US$16.3 billion in the first half of 2026
Nearly US$17 billion in the September quarter alone
In other words, a single recent quarter brought in more than the first six months of the year, and the first half was more than three times the whole of the previous year.
It isn't yet consistently profitable. It reported an operating loss of US$693 million for the first half, and a smaller operating loss in the September quarter.
Why does the Anthropic IPO matter to Australians?
Paul sees this as a test of how much investors are willing to pay for a business built entirely around AI.
Until now, the AI trade has mostly run through companies that existed before AI. Nvidia sold chips for gaming and crypto long before chatbots. Microsoft, Google and Amazon have AI businesses sitting inside much larger ones.
Anthropic is different. It was founded in 2021, it only does AI, and it builds one of the leading frontier models.
A strong debut could lift confidence in the chipmakers, cloud providers and data centre operators tied to AI. A weak one could make investors more cautious about what those businesses are worth.
That matters here because, in Paul's view, AI-related companies have done much of the heavy lifting in global share returns over the past year. Australian shares, which are dominated by banks and miners, have had far less exposure to that theme.
So even someone who never owns an Anthropic share may feel the effect of this float in their super balance over the following months.
Will my super or index fund end up owning Anthropic shares?
Possibly. It depends on what your fund owns and which indexes it follows. Two people can be with the same super fund and hold different investments because they’re in different options. At Guidance, we use broad index funds at the core of client portfolios. When an index adds a company, funds that track it can pick up shares as part of their usual changes.
The timing depends on the index:
Nasdaq-100: under a rule introduced this year, a new listing large enough to rank in the top 40 can be added after 15 trading days.
MSCI World: MSCI has an early-inclusion process for very large new listings, which Paul understands would bring Anthropic in within about 10 trading days. [VERIFY]
S&P 500: it generally requires 12 months of trading and positive earnings over the latest quarter and the past four quarters combined. Anthropic is unlikely to qualify for some time.
For a diversified investor, the direct holding would likely be a small slice. The more meaningful point is how much AI exposure a portfolio already has through the companies that are already in those indexes. We looked at that overlap in Is AI a Bubble? What It Means for Your Investments.
Why is Anthropic going public if it doesn't need the money?
This was the first thing that jumped out at me.
As at 30 June, Anthropic had about US$132 billion in cash and short-term investments. Over its whole life it has raised about US$141 billion. So most of the money it has ever raised is still on the balance sheet.
Part of the reason is that it has paid for much of its computing power with shares rather than cash.
So why list? The most likely answer is that it gives early investors, founders and staff a way to sell. Their shares are released in stages rather than all at once, but over roughly the first year, existing holders get the option to cash out.
A float gives existing shareholders a way to sell over time. The offer documents can show how much money goes to the company through new shares and how much goes to the shareholders selling.
How is Anthropic's ownership structure different?
Buying shares gives investors an ownership stake. Anthropic’s proposed structure gives the founders and a separate trust distinct roles in how the company is run.
It's a public benefit corporation. Under US law, the directors of a normal company are expected to act in shareholders' financial interests. A public benefit corporation's board also has to weigh a stated purpose, which in Anthropic's case is the responsible development of advanced AI for the long-term benefit of humanity.
A separate trust elects most of the board. The Long-Term Benefit Trust has five financially independent trustees chosen for expertise in areas like AI safety and public policy. It generally elects a majority of directors and has no economic interest in the company.
The founders keep voting control. An entity owned by the seven co-founders will hold 50.1% of the vote on most matters, even though they'll own well under half the shares. Public shareholders, together, will have less than half the vote.
Paul compared this with OpenAI, where the board's attempt to remove Sam Altman ended with the board backing down. Anthropic has gone the other way and built its safety-first structure into the listed company itself.
For public shareholders, that means the board weighs Anthropic’s stated public benefit alongside financial results, while the founders retain majority voting power on most matters.
What's the bull case for Anthropic?
I keep coming back to the way businesses use Claude. A team might try it for one job and end up using it every day.
The demand is real. The revenue figures show how much customers are already spending. AI is turning up everywhere, from super fund websites to retailer chatbots.
It sells to businesses. Think about the Excel spreadsheet someone built years ago that the whole team still relies on. I can see Claude becoming that familiar in businesses that build their own tools around it. More than 11,900 customers spend US$100,000 or more a year with Anthropic, and more than 1,900 spend over US$1 million. Even large enterprise customers are still spending a small share of their software budgets on it, which leaves room to grow.
Its backers want it to succeed. Google and Amazon are both significant shareholders as well as cloud suppliers. Google competes with Anthropic through Gemini, but it also has a stake in Anthropic doing well.
Coding is a high-value use. Claude is known for writing software, which is one of the clearest places AI can save expensive time.
People pay for it. Unlike a Google search, many AI tools charge a monthly subscription. If adoption keeps spreading, that's a lot of recurring revenue.
On the founders cashing out, my reply was that early exits don't doom a company. Paul Allen left Microsoft early and bought himself a football team and a superyacht. Microsoft did fine.
What's the bear case for Anthropic?
Paul’s case came down to what Anthropic has to spend to keep customers and how easily those customers could take their work elsewhere.
Who's selling, and why? Insiders know the business best. If the float exists mainly so they can sell, public investors should ask whether they're the ones left holding the baby.
A lot of good news is priced in. At US$1.5 to 2 trillion, Anthropic would be valued at roughly 22 to 30 times its current annualised revenue. Paul's rough comparison put that above where Google and Facebook were priced when they floated, though below SpaceX's float earlier this year. [VERIFY] A good company can still be a poor investment at the wrong price.
Infrastructure is a new skill. Anthropic has mostly rented computing power from Amazon, Google and Microsoft. It's now leasing hardware and planning data centres of its own. Its amended prospectus discloses around US$895 billion of future compute and infrastructure commitments, nearly half of it expected by the end of 2030. Building model software and running data centres are very different jobs.
More usage can mean a bigger computing bill. Think of someone on a flat monthly plan who starts giving Claude twice as much work. Their fee stays the same while Anthropic has more work to process. The cost per request may come down over time. This was the argument that stumped me for a minute.
Customers may have cheaper options. This was Paul’s strongest point. Chinese developers are releasing models that aren't far behind, often without the newest chips, and some are open for businesses to adapt. He compared it to electric cars: Tesla led early, then cheaper Chinese competitors caught up. If AI models become interchangeable, it may not be a great business to be in, especially with huge infrastructure bills attached.
My answer to the last one was Apple, which still dominates despite plenty of competition. Paul's reply: how do we know Anthropic isn't the next BlackBerry?
Fair point, though Anthropic does have something its larger competitors don't. It does only one thing, so all its attention goes there.
Is the Anthropic IPO another dot-com bubble?
Paul is firm that it isn't, at least not in the dot-com sense.
The late-1990s boom was full of companies with fast-rising share prices and little or no revenue. Anthropic took nearly US$17 billion in a single quarter. That's a very different business.
Paul thinks the rollout of electricity is a better comparison. He also brought up Telstra. When your phone still has a signal halfway down a country road, it’s easy to see what the network is worth to you. The company has to work out whether enough people will pay for that coverage to cover the cost of building and running it. Paul made a similar point about America’s railways. Many of the companies that built them went broke.
The same question applies to AI data centres. How much revenue will they bring in over their lifetime compared with what they cost to build and run?
How could the Anthropic float affect existing investments?
A global index fund could gain a small Anthropic holding if its index adds the company. Changes in how investors value AI businesses could have a wider effect. If the shares halve in the week after listing, that could ripple across the AI-linked companies that already make up a big part of global markets. If the float goes well, the effect could run the other way.
Either way, it may show up in super balances over the following six to twelve months.
For some Australian super and ETF investors, the bigger connection may be the companies they already own through global share funds.
Want help with your investments
Whether you’re building wealth or getting closer to retirement, we can help you with your investment strategy. Book an appointment below.
Common questions
When is the Anthropic IPO?
As at 4 November 2026, no date has been confirmed. Reports suggest mid to late November 2026, after the US midterm elections. The shares are expected to trade on the Nasdaq under the ticker ANT.
Can Australians buy Anthropic shares?
Once listed, Anthropic shares would generally be available through Australian platforms that offer US shares. Access to the float itself depends on the terms of the proposed Australian retail offer. Anthropic’s list of expected Australian brokers
Will my super fund own Anthropic?
Possibly. Many super funds invest in global shares by tracking broad indexes, and very large new listings tend to be added to those indexes over time. The Nasdaq-100 can add a qualifying new listing after 15 trading days, while the S&P 500 generally requires a year of trading and a record of profits.
What is Anthropic's Long-Term Benefit Trust?
It's a body of five financially independent trustees that generally elects a majority of Anthropic's board. It has no economic interest in the company. Its role is to keep the company focused on its public benefit purpose as well as its financial results.
Is Anthropic profitable?
Not consistently. It reported an operating loss of US$693 million in the first half of 2026 and a smaller operating loss in the September quarter, while revenue grew very quickly.
Nick Donato is a financial adviser at Guidance Financial Services in Essendon, Melbourne, helping Australians plan for retirement.
Get GainingCHOICE in your inbox.
Each week, we unpack the money headlines, financial ideas and real-life questions that can help you feel more confident with your next financial move. Sign up below:
This article is for educational purposes only and does not take into account your individual circumstances. If you would like tailored advice, we can help you work through the numbers properly.
