When Mike Cannon Brooks and Scott Farquhar rang the Nasdaq bell in December 2015, it seemed like the beginning of something inevitable.
Their company, Atlassian, was more than just a software float, it was a proving point for Australia. It means companies founded in Sydney can sit at the same table as the best companies in Silicon Valley.
The company’s business model of selling software subscriptions seemed like the future at a time when manufacturing things seemed a bit like the past.
Ten years later, the world has turned upside down. Atlassian will be removed from the Nasdaq 100 (a list of the top 100 companies listed on major tech exchanges) and replaced by SanDisk, a hardware maker known for memory cards and flash drives long before the dot-com bubble and the iPhone.
This ouster is a symbolic handover, showing how brutally AI is disrupting the software sector.
SanDisk enters with a market capitalization of about $100 billion (approximately $140 billion). Atlassian is exiting the company for about $17 billion. Just one year ago, SanDisk stock was trading at nearly $33. Current price is approximately $905. That’s a 2,700% run, turning a $35,000 punt into $1 million. Despite reporting its first $1 billion cloud quarter, Atlassian has lost about two-thirds of its value over the same period, with its stock price down 85% from its 2021 peak.
But that doesn’t mean Atlassian’s business is doomed. By most operational measures, it’s the most powerful yet. What’s broken is the theory that investors spent 10 years convincing themselves of.
Atlassian was supposed to embody the era of enterprise software, built around recurring revenue from software that costs little to distribute to more users. But the market has decided that even strong software franchises can become vulnerable if AI changes the economics of coding.
The wildest version of the anti-Atlassian argument is that investors are now worried that AI will make software easier to replicate, while at the same time eliminating many of the software engineering roles that use it. That’s why the stock price is being punished, even though the company has strong earnings.
People briefed on Atlassian’s biggest customers and backers say the company’s bull case is simple: There’s time. Atlassian’s most important customers are large vessels that require long turnaround times.
If a major retailer or financial institution wants to change its software, it can take years and cost billions of dollars. Just ask companies like the ASX and ANZ that have pumped billions into technology projects. Therefore, when many large companies consider changing technology tools and consider vendor offerings based on functionality, they often award incumbents twice as many points as new companies to account for the difficulty of switching.
This is an advantage that can protect Atlassian from being abandoned by its largest customers for 18-36 months. Meanwhile, Atlassian isn’t standing still either. The company has been quietly acquiring AI companies, including Secoda, a little-known company that helps companies use AI to answer questions from their proprietary data.
“People talk to Atlassian [another pre-AI big tech company] Salesforce needs to catch up,” said one of the people briefed on the market position.
Cannon-Brookes makes a similar argument. Last week, he called claims of “vibecoding your own software” among serious buyers “nonsense” and pointed out that developers make up just 40% of Atlassian users anyway.
SanDisk, by contrast, doesn’t need to tell a story. All you need to do is ship the storage. Every data center where you train a Frontier model needs what Nvidia’s Jensen Huang called “the world’s AI working memory”: storage for AI models to retrieve in real time.
The semiconductor company has increased its “price per bit” by more than 100% over the past year and cannot manufacture products fast enough.
This kind of growth has brought opportunists. Last Wednesday, Allbirds, the shoe company that symbolized Obama-era Silicon Valley with its wool running shoes (really), its responsible business pledge, and its $4 billion initial public offering, announced it was selling itself to a brand licensing company and reinventing its remaining shell as “NewBird AI.” Backed by a $50 million investment, the company plans to purchase graphics processing units (GPUs) and lease them to developers.
The stock, which had soared before the announcement, rose 600% in one day. By Thursday, it had already fallen 36%.
We are firmly in the euphoric phase of the hardware cycle when distressed wool runner companies can buy up computer chips and enrich their stocks.
Although the swap from Atlassian to SanDisk and the pivot from Allbirds to Newbird seem like opposites, they are actually the same thing. A 40-year-old storage company and a GPU-buying zombie sneaker brand now occupy the same narrative slot that an enterprise software company held just a year ago.
Cannon and Brooks may be proven right within five years, and history may look back on this exchange as the exact moment the market temporarily lost its mind. But the hierarchy is changing, at least for now. Investors want either direct use of AI or critical infrastructure roles. Everything else is yesterday’s news.
Our Business Briefing newsletter brings you top stories, exclusive coverage and expert opinion. Sign up to receive it every weekday morning..
David Swan is technology editor at The Age and The Sydney Morning Herald. He was previously technology editor at The Australian Newspapers.Connect via X or Email.
Nick Bonyhady is business editor of The Sydney Morning Herald and The Age. He is a former federal deputy editor, technology editor, and labor relations reporter.Connect via X or Email.
