The broader AI market is being called a bubble by some and a revolution by others. But there is a separate question worth asking specifically about Meta: is this stock, right now, a bubble of its own making, inflated by hype around an AI bet that has yet to show a measurable intelligence of its own?&nbsp;The Capex Number That Shocked Wall Street&nbsp;When Meta reported its Q1 2026 earnings, the headline revenue numbers were strong, $56.31 billion in quarterly revenue, a 33% year-on-year jump, with advertising revenue of $55.02 billion driving almost all of it. But what moved markets was something else entirely: Meta announced it was raising its 2026 capital expenditure forecast to between $125 billion and $145 billion. That is up from its January guidance of $115 to $135 billion, and represents an increase of over 60% compared to what it spent the previous year. Meta's stock fell 8 to 10% in early trading the following day. Investors were not celebrating the revenue beat. They were staring at a capex number that is, by any measure, staggering.&nbsp;The Human Cost Nobody Is Pricing Into Meta's Stock&nbsp;There is another dimension to Meta's AI story that Wall Street is largely ignoring: the catastrophic real-world consequences of deploying a cheap, low-quality AI system at billion-user scale with essentially no human oversight left to catch its errors. Meta has systematically laid off its human content moderation teams, the people who used to review appeals, reinstate wrongly banned accounts, and apply judgment to edge cases. What has replaced them is an AI system that creators, journalists, and small business owners across the world describe as erratic, opaque, and unaccountable. Accounts with years of original content, verified identities, and legitimate income streams are being terminated or restricted without warning and without any realistic path to appeal, while some reporting that fake accounts with blue tick verification badges continue to operate freely on the same platforms. The engineers building this system, referred to internally in some circles as working in conditions resembling a content gulag, are reportedly producing training tasks so rudimentary they resemble puzzles designed for toddlers. That is the training data feeding an AI that is now making consequential, life-altering decisions about whether a creator's account survives, whether a small business keeps its primary sales channel, whether a journalist's archive disappears overnight. The gap between the simplicity of what is going into Meta AI and the seriousness of what it is doing to real people is not a technical footnote. It is a liability, legal, reputational, and financial, that does not yet appear anywhere in Meta's $145 billion capex plan.&nbsp;Spending More Than It Earns On AI, With No Cloud To Show For It&nbsp;The core problem with Meta's AI spending is structural. Google, Amazon, and Microsoft are also pouring hundreds of billions into AI infrastructure, but each of them has a cloud computing business that can immediately monetise that infrastructure by selling compute capacity to other companies. Meta has no such business. Its revenue is almost entirely advertising. That means every dollar it spends building data centres and buying Nvidia chips is a bet that AI will eventually make its ads more targeted, its platforms more engaging, and its users more sticky. That may well happen. But right now, it is a hypothesis, not a revenue line.&nbsp;The Zuckerberg ROI Quote That Should Concern Investors&nbsp;When Meta's Q1 2026 earnings call touched on what all this AI spending was actually delivering in terms of return on investment, Mark Zuckerberg's response was notable. Asked about measurable ROI from the AI build-out, Zuckerberg reportedly described it as "a very technical question." For a company spending $145 billion in a single year on capital projects, that answer is not reassuring. And the earnings quality underneath the headline numbers is softer than it appears: Meta's reported GAAP EPS of $10.44 was significantly inflated by a one-time $8.03 billion tax benefit linked to the Trump administration's tax policy. Strip that out and the underlying EPS was approximately $7.31, a number that tells a much more cautious story about actual profitability growth.&nbsp;Michael Burry's Warning&nbsp;The investor made famous by The Big Short, Michael Burry, has specifically flagged Big Tech's accounting treatment of AI hardware as a concern. His argument is that companies including Meta are extending the stated useful life of their computing assets, primarily Nvidia chips, which reduces annual depreciation charges and makes earnings look better than they are. Burry's estimates suggest Meta could be overstating its earnings by as much as 20.8% by 2028 if this accounting treatment continues. Whether or not that proves correct, the concern highlights that Meta's profitability picture may be softer than its headline numbers suggest.&nbsp;Where The Stock Stands&nbsp;Meta's stock hit an all-time high of $796.25 per share in August 2025. As of mid-June 2026, it is trading at $577.22, a decline of approximately 27.5% from that peak. At a market capitalisation of around $1.46 trillion and a business that remains almost entirely dependent on advertising revenue, Meta is not cheaply priced for a company whose AI bet has yet to prove itself operationally as well ass financially. The AI market at large may or may not be a bubble. But Meta's specific combination of enormous spending, advertising-only revenue, unproven AI returns, and a moderation crisis affecting millions of real users makes it a stock with a bubble story entirely its own.

Meta is spending $145 billion on AI in 2026 with no cloud business to monetise it. Its stock is down 21% from its ATH. Here's why Meta could be a bubble all on its own.

The broader AI market is being called a bubble by some and a revolution by others. But there is a separate question worth asking specifically about Meta: is this stock, right now, a bubble of its own making, inflated by hype around an AI bet that has yet to show a measurable intelligence of its own? The Capex Number That Shocked Wall Street When Meta reported its Q1 2026 earnings, the headline revenue numbers were strong, $56.31 billion in quarterly revenue, a 33% year-on-year jump, with advertising revenue of $55.02 billion driving almost all of it. But what moved markets was something else entirely: Meta announced it was raising its 2026 capital expenditure forecast to between $125 billion and $145 billion. That is up from its January guidance of $115 to $135 billion, and represents an increase of over 60% compared to what it spent the previous year. Meta's stock fell 8 to 10% in early trading the following day. Investors were not celebrating the revenue beat. They were staring at a capex number that is, by any measure, staggering. The Human Cost Nobody Is Pricing Into Meta's Stock There is another dimension to Meta's AI story that Wall Street is largely ignoring: the catastrophic real-world consequences of deploying a cheap, low-quality AI system at billion-user scale with essentially no human oversight left to catch its errors. Meta has systematically laid off its human content moderation teams, the people who used to review appeals, reinstate wrongly banned accounts, and apply judgment to edge cases. What has replaced them is an AI system that creators, journalists, and small business owners across the world describe as erratic, opaque, and unaccountable. Accounts with years of original content, verified identities, and legitimate income streams are being terminated or restricted without warning and without any realistic path to appeal, while some reporting that fake accounts with blue tick verification badges continue to operate freely on the same platforms. The engineers building this system, referred to internally in some circles as working in conditions resembling a content gulag, are reportedly producing training tasks so rudimentary they resemble puzzles designed for toddlers. That is the training data feeding an AI that is now making consequential, life-altering decisions about whether a creator's account survives, whether a small business keeps its primary sales channel, whether a journalist's archive disappears overnight. The gap between the simplicity of what is going into Meta AI and the seriousness of what it is doing to real people is not a technical footnote. It is a liability, legal, reputational, and financial, that does not yet appear anywhere in Meta's $145 billion capex plan. Spending More Than It Earns On AI, With No Cloud To Show For It The core problem with Meta's AI spending is structural. Google, Amazon, and Microsoft are also pouring hundreds of billions into AI infrastructure, but each of them has a cloud computing business that can immediately monetise that infrastructure by selling compute capacity to other companies. Meta has no such business. Its revenue is almost entirely advertising. That means every dollar it spends building data centres and buying Nvidia chips is a bet that AI will eventually make its ads more targeted, its platforms more engaging, and its users more sticky. That may well happen. But right now, it is a hypothesis, not a revenue line. The Zuckerberg ROI Quote That Should Concern Investors When Meta's Q1 2026 earnings call touched on what all this AI spending was actually delivering in terms of return on investment, Mark Zuckerberg's response was notable. Asked about measurable ROI from the AI build-out, Zuckerberg reportedly described it as "a very technical question." For a company spending $145 billion in a single year on capital projects, that answer is not reassuring. And the earnings quality underneath the headline numbers is softer than it appears: Meta's reported GAAP EPS of $10.44 was significantly inflated by a one-time $8.03 billion tax benefit linked to the Trump administration's tax policy. Strip that out and the underlying EPS was approximately $7.31, a number that tells a much more cautious story about actual profitability growth. Michael Burry's Warning The investor made famous by The Big Short, Michael Burry, has specifically flagged Big Tech's accounting treatment of AI hardware as a concern. His argument is that companies including Meta are extending the stated useful life of their computing assets, primarily Nvidia chips, which reduces annual depreciation charges and makes earnings look better than they are. Burry's estimates suggest Meta could be overstating its earnings by as much as 20.8% by 2028 if this accounting treatment continues. Whether or not that proves correct, the concern highlights that Meta's profitability picture may be softer than its headline numbers suggest. Where The Stock Stands Meta's stock hit an all-time high of $796.25 per share in August 2025. As of mid-June 2026, it is trading at $577.22, a decline of approximately 27.5% from that peak. At a market capitalisation of around $1.46 trillion and a business that remains almost entirely dependent on advertising revenue, Meta is not cheaply priced for a company whose AI bet has yet to prove itself operationally as well ass financially. The AI market at large may or may not be a bubble. But Meta's specific combination of enormous spending, advertising-only revenue, unproven AI returns, and a moderation crisis affecting millions of real users makes it a stock with a bubble story entirely its own.

Meta is spending up to $145 billion on AI infrastructure in 2026 alone. Its stock has already dropped 21% from its all-time high. Revenue is growing fast but profits are not keeping pace.&nbsp;

5 reasons why 'Meta' could be the BIGGEST BUBBLE of AI era

Artificial intelligence is fueling one of the biggest investment waves in modern history, with AI-focused companies attracting staggering valuations and reshaping global markets.

But as billions pour into the sector and investors chase the next AI giant, questions are growing over whether the industry is witnessing a genuine technological revolution—or the formation of a massive financial bubble. Is the AI boom sustainable, or are markets getting ahead of reality?

Trillion-dollar AI Boom: Bubble or Breakthrough?

Jeff Bezos has addressed growing concerns around a possible artificial intelligence bubble, saying that even if such a situation occurs, it should not be a cause for worry. Speaking about the rapid growth in AI investments, Bezos suggested that periods of hype and over-investment are often part of how new technologies develop and mature.<h2>What Bezos said about an AI bubble</h2>Bezos acknowledged that some experts believe the current surge in AI spending could lead to a bubble, similar to past technology cycles. However, he argued that even if investment exceeds short-term returns, it still helps build long-term infrastructure and innovation.According to him, such phases often result in:Better technology development Stronger infrastructure Wider adoption of new tools This means that even if a correction happens, the overall progress of AI will continue.<h2>Why AI investment is rising rapidly</h2>The artificial intelligence sector has seen significant growth in recent years. Companies are investing billions of dollars into:Data centres and computing power AI models and research Software and enterprise toolsIndustry reports show that global AI spending is expected to cross hundreds of billions of dollars in the coming years, reflecting strong demand from businesses and governments. This rapid growth has led to comparisons with earlier technology booms, including the dot-com era.<h2>Lessons from past tech bubbles</h2>Bezos pointed out that past bubbles, such as the internet boom in the late 1990s, eventually led to lasting innovation despite short-term market corrections. While many companies did not survive those periods, the underlying technology continued to grow and shape the modern economy.<h2>Experts say a similar pattern could occur with AI:</h2>Some companies may struggle Investments may slow down But core technology will continue to improve&nbsp;<h2>Why experts remain divided</h2> There is ongoing debate among analysts about whether AI is currently in a bubble phase.Some argue that:Valuations of AI companies are rising quickly Spending on infrastructure may be ahead of demand Others believe that:AI adoption is still in early stages Long-term use cases will justify current investments Bezos’ comments align with the view that long-term benefits outweigh short-term risks.<h2>What this means for businesses and investors</h2> For businesses, the focus remains on adopting AI tools to improve efficiency and productivity. For investors, the discussion highlights the importance of understanding both risks and opportunities in emerging technologies. Even if markets fluctuate, companies that build strong foundations in AI may benefit over time. The future of AI investment will depend on how quickly real-world applications grow and generate returns. While concerns about a bubble may continue, industry leaders like Bezos suggest that the long-term impact of AI is likely to remain strong. The debate reflects a broader question facing the tech industry: how to balance rapid innovation with sustainable growth

Jeff Bezos has addressed growing concerns around a possible artificial intelligence bubble, saying that even if such a situation occurs, it should not be a cause for worry. 

Jeff Bezos has addressed growing concerns around a possible artificial intelligence bubble, saying that even if such a situation occurs, it should not be a cause for worry. Speaking about the rapid growth in AI investments, Bezos suggested that periods of hype and over-investment are often part of how new technologies develop and mature. What Bezos said about an AI bubble Bezos acknowledged that some experts believe the current surge in AI spending could lead to a bubble, similar to past technology cycles. However, he argued that even if investment exceeds short-term returns, it still helps build long-term infrastructure and innovation. According to him, such phases often result in: Better technology development Stronger infrastructure Wider adoption of new tools This means that even if a correction happens, the overall progress of AI will continue. Why AI investment is rising rapidly The artificial intelligence sector has seen significant growth in recent years. Companies are investing billions of dollars into: Data centres and computing power AI models and research Software and enterprise tools Industry reports show that global AI spending is expected to cross hundreds of billions of dollars in the coming years, reflecting strong demand from businesses and governments. This rapid growth has led to comparisons with earlier technology booms, including the dot-com era. Lessons from past tech bubbles Bezos pointed out that past bubbles, such as the internet boom in the late 1990s, eventually led to lasting innovation despite short-term market corrections. While many companies did not survive those periods, the underlying technology continued to grow and shape the modern economy. Experts say a similar pattern could occur with AI: Some companies may struggle Investments may slow down But core technology will continue to improve Why experts remain divided There is ongoing debate among analysts about whether AI is currently in a bubble phase. Some argue that: Valuations of AI companies are rising quickly Spending on infrastructure may be ahead of demand Others believe that: AI adoption is still in early stages Long-term use cases will justify current investments Bezos’ comments align with the view that long-term benefits outweigh short-term risks. What this means for businesses and investors For businesses, the focus remains on adopting AI tools to improve efficiency and productivity. For investors, the discussion highlights the importance of understanding both risks and opportunities in emerging technologies. Even if markets fluctuate, companies that build strong foundations in AI may benefit over time. The future of AI investment will depend on how quickly real-world applications grow and generate returns. While concerns about a bubble may continue, industry leaders like Bezos suggest that the long-term impact of AI is likely to remain strong. The debate reflects a broader question facing the tech industry: how to balance rapid innovation with sustainable growth

Jeff Bezos has addressed growing concerns around a possible artificial intelligence bubble, saying that even if such a situation occurs, it should not be a cause for worry.&nbsp;

AI boom or bubble? Jeff Bezos offers a different perspective

Larry Fink, the CEO of US financial giant and the world's largest asset manager firm BlackRock, has warned that if oil prices reach $150 a barrel, it will trigger a global recession. According to him, “If Iran remains a threat and oil prices stay high, it will have profound implications for the world economy”.&nbsp;Fink predicted two extreme scenarios, one in which the conflict is settled, and Iran is integrated into the global economy, and then the prices of oil will fall back to where they were before the war. If it were not, the prices of oil would rise to “years of above $100, closer to $150 oil” and an outcome of "a probably stark and steep recession". "Rising energy prices are a very regressive tax. It affects the poor more than the wealthy," said Fink.&nbsp;Also Read: <a href="https://www.wionews.com/world/blackrock-hlend-fund-withdrawal-limit-private-credit-1772868211906">Liquidity Squeeze: BlackRock caps withdrawals from $26B HLEND fund amid investor anxiety</a>In an interview with the BBC, Fink was seemingly optimistic about the speculation of AI development and asserted that there are “zero” similarities with the financial crisis led by the ‘real-estate’ bubble. This is despite the large narrative driven investement in AI infrastructure, no proper revenue model and huge speculative trading to inflate the AI market.&nbsp;As energy prices are surging, there are hints of cracks in the base of the global economy, which is “the petrodollar system”. BlackRock itself have restricted withdrawal from investors as the liquidity crunch is quickly approaching. Some fringe analysts are also predicting that the US war of aggression in Iran is a “controlled demolition” of the old world order to a new world order driven by “AI and surveillance”.&nbsp;"I do not believe we have a bubble at all," he says. "Could we have one or two failures in AI? Sure, that I'm fine with." In 2025, Black Rock was part of the consortium that bought one of the world's largest data centre providers, Aligned Data Centres, for a $40 million deal. So "there is no bubble" is hard to swallow pill from someone whose firm will be badly hit if there were a bubble.&nbsp;

BlackRock CEO Larry Fink warns that oil prices nearing $150 could trigger a global recession, citing risks from Iran tensions and rising energy costs.

Larry Fink, the CEO of US financial giant and the world's largest asset manager firm BlackRock, has warned that if oil prices reach $150 a barrel, it will trigger a global recession. According to him, “If Iran remains a threat and oil prices stay high, it will have profound implications for the world economy”. Fink predicted two extreme scenarios, one in which the conflict is settled, and Iran is integrated into the global economy, and then the prices of oil will fall back to where they were before the war. If it were not, the prices of oil would rise to “years of above $100, closer to $150 oil” and an outcome of "a probably stark and steep recession". "Rising energy prices are a very regressive tax. It affects the poor more than the wealthy," said Fink. Also Read: Liquidity Squeeze: BlackRock caps withdrawals from $26B HLEND fund amid investor anxiety In an interview with the BBC, Fink was seemingly optimistic about the speculation of AI development and asserted that there are “zero” similarities with the financial crisis led by the ‘real-estate’ bubble. This is despite the large narrative driven investement in AI infrastructure, no proper revenue model and huge speculative trading to inflate the AI market. As energy prices are surging, there are hints of cracks in the base of the global economy, which is “the petrodollar system”. BlackRock itself have restricted withdrawal from investors as the liquidity crunch is quickly approaching. Some fringe analysts are also predicting that the US war of aggression in Iran is a “controlled demolition” of the old world order to a new world order driven by “AI and surveillance”. "I do not believe we have a bubble at all," he says. "Could we have one or two failures in AI? Sure, that I'm fine with." In 2025, Black Rock was part of the consortium that bought one of the world's largest data centre providers, Aligned Data Centres, for a $40 million deal. So "there is no bubble" is hard to swallow pill from someone whose firm will be badly hit if there were a bubble.

BlackRock CEO Larry Fink warns that oil prices nearing $150 could trigger a global recession, citing risks from Iran tensions and rising energy costs.

BlackRock CEO Larry Fink warns $150 oil could trigger global recession

Despite the debate that Artificial Intelligence firms are in a bubble that would burst soon, several of them are preparing for IPOs. In the US, the contest is mainly between OpenAI and its closest rivals. In China, some big names in the industry, collectively known as ‘AI Tigers’, have either already listed or are in the process. Billions, and possibly trillions, of dollars are at stake. Here is an update.<h2>OpenAI IPO: Plans to beat Anthropic</h2>OpenAI, the company behind the popular chatbot ChatGPT, wants to go public by the last quarter of 2026 and is trying to beat Anthropic to become the first major generative AI startup to IPO in the US. Valued at around $830 billion, which could rise to $1 trillion at listing, OpenAI has been on a fundraising spree. Major funders include Microsoft and SoftBank, with possible involvement from Amazon as well. This is despite OpenAI projecting cumulative losses of $115 billion through 2029, against annual revenue projections of around $20 billion.<h2>Anthropic is readying for an IPO</h2>Anthropic, the company behind the Claude AI models, is planning an IPO by mid-2026. Its valuation is around $350 billion following a $10 bn funding round. Anthropic’s investors include Microsoft, Nvidia and Amazon. Revenue is projected to reach $26 billion this year.<a href="https://www.wionews.com/videos/south-korea-enacts-ai-act-after-deepfake-menance-shocks-world-1769745050466/embed">https://www.wionews.com/videos/south-korea-enacts-ai-act-after-deepfake-menance-shocks-world-1769745050466/embed</a><h2>Musk wants an xAI IPO along with SpaceX: The mother of all AI IPOs?</h2>Elon Musk’s AI company xAI, which is behind the controversial Grok chatbot, is in merger discussions with SpaceX, his space exploration company. The plan appears to be a combined IPO in mid-2026. The combined entity could be valued at $1 trillion. However, xAI is valued at a much lower $230 billion as a standalone company. The AI firm has recently raised $20 billion in a Series E round and has a $2 billion funding commitment from SpaceX. Nvidia and Cisco are among its strategic investors.Also read: <a href="https://www.wionews.com/technology/ai-generated-content-will-shape-the-future-of-social-media-says-mark-zuckerberg-1769668052622">AI-generated content will shape the future of social media, says Mark Zuckerberg</a><h2>Databricks, Cerebras Systems, Cohere and other potential IPO candidates</h2>AI infrastructure provider Databricks could IPO, with its valuation having reached $134 billion in 2025. The company is backed by Nvidia, Microsoft and Amazon, with a revenue run rate of around $4.8 billion.AI chipmaker Cerebras Systems, which is focused on high-performance GPUs, could IPO in the second quarter of 2026. Valued at around $8.1 billion, Cerebras has investors including Fidelity, Sequoia Capital, Foundation Capital, Eclipse, Coatue and Benchmark.The enterprise-focused AI model developer Cohere is another IPO candidate this year, with a $7 billion valuation and $150 million in annual recurring revenue.Groq, the company that makes AI chips (and is unrelated to Grok), and AI robotics firm Figure AI are also holding pre-IPO discussions, without confirmed IPO dates.Also read: <a href="https://www.wionews.com/photos/-120-billion-ai-bubble-at-risk-ibm-ceo-says-ai-bubble-could-burst-soon-1769682978130">$120 billion AI bubble at risk? IBM CEO says AI bubble could burst soon</a><h2>Meanwhile in China: Another AI IPO boom is underway, with some already listed</h2>China is the other major market seeing a scramble for AI IPOs, led by generative “AI tigers” and AI chip companies. Some have already gone public, while others are preparing to IPO.Among the first to go public was Zhipu AI, a leading “AI tiger” that developed the GLM family of large language models (LLMs). It became the world’s first major LLM company to go public on January 8, with shares closing 13.2 per cent higher on debut. The company is valued at approximately $6.7 billion post-IPO.MiniMax, another “AI tiger”, known for video generation as well as multimodal AI across text, image and voice, completed its Hong Kong IPO on January 9. Shares of the company behind Talkie AI and Hailuo AI surged 109 per cent on debut, with a current post-IPO valuation of around $13.7 billion. Alibaba is its largest investor, while other backers include Tencent, Hillhouse Capital, Future Capital, Xiaomi, Xiaohongshu, Shunwei and IDG.Moonshot AI, which is behind the Kimi chatbot, could IPO later in 2026.Major Chinese semiconductor companies are also going public. Kunlunxin, the AI chip unit of Baidu, which is developing chips to rival Nvidia, has plans for a Hong Kong IPO in 2026.&nbsp;T-Head, Alibaba’s chip unit, could pursue a spinoff IPO. Montage Technology, another AI chip designer for data centres and servers, is pursuing a secondary listing on the Hong Kong Exchange in February. Axera Semiconductor, a maker of AI inference chips, is also set to list in Hong Kong in February

Despite talk of a bubble, AI firms are rushing towards IPOs in the US and China. In the US, the race is between OpenAI, and its competitors like Anthropic and xAI. In China, AI Tigers' and chipmakers are listing fast with big valuations

Despite the debate that Artificial Intelligence firms are in a bubble that would burst soon, several of them are preparing for IPOs. In the US, the contest is mainly between OpenAI and its closest rivals. In China, some big names in the industry, collectively known as ‘AI Tigers’, have either already listed or are in the process. Billions, and possibly trillions, of dollars are at stake. Here is an update. OpenAI IPO: Plans to beat Anthropic OpenAI, the company behind the popular chatbot ChatGPT, wants to go public by the last quarter of 2026 and is trying to beat Anthropic to become the first major generative AI startup to IPO in the US. Valued at around $830 billion, which could rise to $1 trillion at listing, OpenAI has been on a fundraising spree. Major funders include Microsoft and SoftBank, with possible involvement from Amazon as well. This is despite OpenAI projecting cumulative losses of $115 billion through 2029, against annual revenue projections of around $20 billion. Anthropic is readying for an IPO Anthropic, the company behind the Claude AI models, is planning an IPO by mid-2026. Its valuation is around $350 billion following a $10 bn funding round. Anthropic’s investors include Microsoft, Nvidia and Amazon. Revenue is projected to reach $26 billion this year. https://www.wionews.com/videos/south-korea-enacts-ai-act-after-deepfake-menance-shocks-world-1769745050466/embed Musk wants an xAI IPO along with SpaceX: The mother of all AI IPOs? Elon Musk’s AI company xAI, which is behind the controversial Grok chatbot, is in merger discussions with SpaceX, his space exploration company. The plan appears to be a combined IPO in mid-2026. The combined entity could be valued at $1 trillion. However, xAI is valued at a much lower $230 billion as a standalone company. The AI firm has recently raised $20 billion in a Series E round and has a $2 billion funding commitment from SpaceX. Nvidia and Cisco are among its strategic investors. Also read: AI-generated content will shape the future of social media, says Mark Zuckerberg Databricks, Cerebras Systems, Cohere and other potential IPO candidates AI infrastructure provider Databricks could IPO, with its valuation having reached $134 billion in 2025. The company is backed by Nvidia, Microsoft and Amazon, with a revenue run rate of around $4.8 billion. AI chipmaker Cerebras Systems, which is focused on high-performance GPUs, could IPO in the second quarter of 2026. Valued at around $8.1 billion, Cerebras has investors including Fidelity, Sequoia Capital, Foundation Capital, Eclipse, Coatue and Benchmark. The enterprise-focused AI model developer Cohere is another IPO candidate this year, with a $7 billion valuation and $150 million in annual recurring revenue. Groq, the company that makes AI chips (and is unrelated to Grok), and AI robotics firm Figure AI are also holding pre-IPO discussions, without confirmed IPO dates. Also read: $120 billion AI bubble at risk? IBM CEO says AI bubble could burst soon Meanwhile in China: Another AI IPO boom is underway, with some already listed China is the other major market seeing a scramble for AI IPOs, led by generative “AI tigers” and AI chip companies. Some have already gone public, while others are preparing to IPO. Among the first to go public was Zhipu AI, a leading “AI tiger” that developed the GLM family of large language models (LLMs). It became the world’s first major LLM company to go public on January 8, with shares closing 13.2 per cent higher on debut. The company is valued at approximately $6.7 billion post-IPO. MiniMax, another “AI tiger”, known for video generation as well as multimodal AI across text, image and voice, completed its Hong Kong IPO on January 9. Shares of the company behind Talkie AI and Hailuo AI surged 109 per cent on debut, with a current post-IPO valuation of around $13.7 billion. Alibaba is its largest investor, while other backers include Tencent, Hillhouse Capital, Future Capital, Xiaomi, Xiaohongshu, Shunwei and IDG. Moonshot AI, which is behind the Kimi chatbot, could IPO later in 2026. Major Chinese semiconductor companies are also going public. Kunlunxin, the AI chip unit of Baidu, which is developing chips to rival Nvidia, has plans for a Hong Kong IPO in 2026. T-Head, Alibaba’s chip unit, could pursue a spinoff IPO. Montage Technology, another AI chip designer for data centres and servers, is pursuing a secondary listing on the Hong Kong Exchange in February. Axera Semiconductor, a maker of AI inference chips, is also set to list in Hong Kong in February

Despite talk of a bubble, AI firms are rushing towards IPOs in the US and China. In the US, the race is between OpenAI, and its competitors like Anthropic and xAI. In China, AI Tigers' and chipmakers are listing fast with big valuations

AI bubble, what bubble? American AI firms are racing for IPOs; some Chinese ‘AI Tigers’ already listed, others in pipeline

According to the report, forward price-to-earnings (P/E) ratios for listed AI companies have actually fallen over the past three years, even as earnings-per-share estimates have more than doubled. 

'AI boom isn't a bubble': JP Morgan report backs market surge with strong fundamentals

Beginning around 2022 and accelerating into 2024–25, artificial intelligence emerged as the primary focus of investors, tech firms and governments alike. 

'AI Bubble vs Dot-Com Crash': 7 data-backed similarities and differences

ai-bubble

Ai Bubble, News on Ai Bubble, stories on Ai Bubble, articles on Ai Bubble, Ai Bubble stories at Wion, Ai Bubble news, Ai Bubble latest news

ai bubble

5 reasons why 'Meta' could be the BIGGEST BUBBLE of AI era

Trillion-dollar AI Boom: Bubble or Breakthrough?

AI boom or bubble? Jeff Bezos offers a different perspective

BlackRock CEO Larry Fink warns $150 oil could trigger global recession

Forget AI bubble? US firms race for IPO; some Chinese ‘Tigers’ already listed

'AI isn't a bubble': JP Morgan report backs market surge with fundamentals

'AI Bubble vs Dot-Com Crash': 7 data-backed similarities and differences