Is spending $600 million on a "concept browser" the biggest sign of the AI bubble?

Recently, the little - known AI startup, The Browser Company, has emerged as the new winner. The enterprise collaboration software giant, Atlassian, acquired it for a staggering $600 million. The most eye - catching asset of the latter is an AI - powered smart browser named Dia, which hasn't been officially launched yet.

A $600 - million cash bet on a "concept" browser

This deal is quite extraordinary. In many technology company mergers and acquisitions, payments are often made with equity of vague value, but Atlassian made a straightforward cash transaction. What's even more interesting is that the product they acquired, the AI - powered smart browser called Dia, hasn't even been officially released to the public. There are only fewer than 100,000 invited users in the internal testing phase. Despite the uncertainties surrounding the product itself, this investment is still considered worthwhile.

Why did Atlassian place such a big bet? They are gambling on the future. They believe that the model demonstrated by Dia, i.e., the seamless integration of AI into people's daily lives, will be a new way for the tech industry to interact with the public. Josh Miller, the co - founder of The Browser Company, said, "The key to this race is to find the equivalent of the 'iPhone app' in this era."

To some extent, the current AI boom bears a resemblance to the dot - com bubble of the past. However, this time, "the numbers seem to be ten times larger, and the money is burning at a faster rate."

Currently, AI products like ChatGPT have hundreds of millions of users, having a substantial impact on the entire US economy. In the second quarter of this year, the US GDP grew by 3%, and nearly half of this growth may be attributed to AI - related capital investments. The construction expenditure on data centers is even expected to exceed that of commercial office buildings for the first time.

However, even staunch supporters of AI admit that a market correction is inevitable, and industry consolidation has already begun, which might be a sign of tightening.

"We're about to enter a 'winter' of company closures," Miller predicted. "In the AI browser market, the winner will be determined within the next 12 months." Ultimately, among these AI startups, only one or two may become the new Apple or Google, while the rest will fade into obscurity.

Building an AI - powered "autopilot" to reshape the way users surf the web

The Dia browser has grand ambitions. It aims to integrate AI interaction into every aspect of the browser, completely changing the way we browse the web.

Currently, AI tools like ChatGPT usually run in separate windows, isolated from our other operations. Dia hopes that whether you're sending and receiving emails, shopping online, or browsing the media, AI chat and generation tools will be readily available at all times. It can help you digest the information you're viewing and transfer it between different tabs.

For example, you can ask Dia to summarize a bunch of Amazon product links you've opened and automatically generate an email. Even more, through an intelligent agent function called "autopilot," it can directly complete the purchase for you.

Miller said that Dia will remember your browsing history. "It will know you very well," and it will integrate text - generation capabilities into the cursor, allowing you to ask questions or get suggestions from the AI at any time.

Tech giants snap up "wizards," and startup founders seek to cash out

Why are big companies so keen on acquiring these startups with uncertain futures? Foreign media, The New Yorker magazine, also analyzed and commented on this:

Firstly, it's a way to shape the future. Google's acquisition of YouTube in 2006 and Facebook's acquisition of Instagram in 2012 have proven that successful deals can set industry trends and attract imitators.

Secondly, in the current environment, acquisitions are a form of marketing. It signals to the market that publicly - traded companies like Atlassian are making progress in the AI field, something they might struggle to achieve on their own.

More importantly, the new - generation AI - native startups possess "black magic" that traditional giants are not familiar with. For instance, traditional programming follows strict logic and is straightforward. Yasmin Razavi, a general partner at Spark Capital, which has invested in major AI startups like Anthropic, said that building AI is more like "brewing beer or kombucha." The same ingredients can produce different results, and the internal operating principles are even a mystery to the operators.

Famous AI scholar Ethan Mollick recently described using AI machines as "dealing with wizards." Since no one can fully understand it, the safest option for tech giants is to recruit those who seem to be experts.

Mike Cannon - Brookes, the co - founder and CEO of Atlassian, recognized this point. He believes that grafting AI features onto traditional software in a clumsy way doesn't work well, and the public is also showing little interest in the AI tools that have been force - fitted into various applications.

Facing this chaotic situation, Josh Stadt, the marketing director of The49, an AI - focused startup incubator, said that many entrepreneurs' goal is not to become the next Zuckerberg but rather "to create something cool and then sell it at a high price and retire." In other words, it might be a wise move to sell their "shovels and picks" at a good price before the bubble bursts.

This AI gold rush continues, and new "shovels" are being sold constantly. But the real question is, who can ultimately strike gold and create a product that goes beyond all the hype and becomes as intuitive as a smartphone? This remains an open question.

This article is from "Tencent Technology," author: Jin Lu, published by 36Kr with authorization.