Anthropic is preparing for what could be the largest IPO in history, targeting a valuation of up to $2 trillion. The maker of Claude AI reported revenue of $16.3 billion in the first half of 2026, a dramatic jump from roughly $1.2 billion a year earlier. However, the company also reported a net loss of $80.7 billion and warned in its filing that advanced AI could pose catastrophic or existential risks to humanity. The S-1 form devoted over 90 pages to risk factors, including the possibility that its technology could help malicious actors develop dangerous weapons.
The five-year-old company plans to list on Nasdaq under the ticker ANT. Five customers accounted for 25 percent of its revenue, and it has signed computing and infrastructure commitments worth over $800 billion through 2045, with deals involving Broadcom, SpaceX, Google, Amazon and Microsoft. If the company reaches its top expected valuation, its founders could each gain as much as $28 billion.
Elsewhere in the AI industry, AMD has agreed to acquire World Labs, an AI research lab led by Dr. Fei-Fei Li, for approximately $8.2 billion in stock. The deal, expected to close by the end of 2026, focuses on spatial-intelligence models that generate 3D environments from text, image and video inputs. Dr. Li will join AMD as executive vice president and chief scientist. The acquisition aims to strengthen AMD's AI hardware, software and systems for an open ecosystem.
Microsoft's security chief Hayete Gallot warned that frontier AI models cannot be trusted, describing AI agents in enterprises as unpredictable agentic insiders that could collaborate to reach objectives. Microsoft recommends using Guardian Agents for real-time monitoring and has developed Project Perception, a system designed to contain agentic insiders with deterministic controls. The warning underscores growing concerns about AI autonomy inside organizations.
Neura Robotics, based in Metzingen, Germany, raised up to $1.4 billion in funding led by Tether and backed by Nvidia and Amazon. The company is building AI-powered robots capable of performing complex human-like tasks, highlighting Europe's efforts to compete in robotics against China. In India, Tata Communications announced a $400 million to $450 million investment over three to five years to build a 10,000-km fibre network called the Indian AI Super Highway, with 7.2 petabytes of capacity connecting major data centres starting with the Mumbai-Pune-Hyderabad-Chennai corridor.
Google is set to report Q3 results with analysts expecting earnings per share of $3.01 on revenue of $127.8 billion. Its cloud unit is projected to see revenue jump 91 percent to $28.9 billion, and the company may earn significant income from selling custom Tensor Processing Units to third-party data centers. TPU sales are expected to grow from $9.4 billion in late 2026 to $138 billion by 2028, though Meta's Muse AI launch poses a potential long-term threat to Google's search dominance.
The broader AI industry continues to struggle with profitability. CoreWeave, the largest neocloud provider, lost 11.7 percent of its stock value in one week after dropping over 25 percent the week before, with its long-term debt interest rate now exceeding 16%. Meanwhile, CrowdStrike trades at a price-to-sales ratio of 49.8x, far above the software industry average of 4.0x and AI peers at about 15.5x, partly due to its partnership with NVIDIA. The American Chamber of Commerce in Korea held a workshop in Seoul urging companies to train workers and manage risks as they adopt AI tools like ChatGPT to gain real productivity gains.
Key Takeaways
- Anthropic filed for an IPO targeting a valuation of up to $2 trillion, with revenue reaching $16.3 billion in the first half of 2026 but a net loss of $80.7 billion.
- Anthropic's IPO filing warns that advanced AI could pose catastrophic or existential risks, including helping malicious actors develop dangerous weapons.
- AMD agreed to acquire World Labs for $8.2 billion in stock, with Dr. Fei-Fei Li joining AMD as executive vice president and chief scientist.
- Microsoft's security chief warns that frontier AI models cannot be trusted and recommends Guardian Agents and Project Perception to monitor AI agents in enterprises.
- Neura Robotics raised $1.4 billion led by Tether, with backing from Nvidia and Amazon, to build human-like AI robots in Germany.
- Tata Communications plans to invest $400-$450 million over three to five years to build a 10,000-km AI fibre network across India with 7.2 petabytes of capacity.
- Google expects Q3 cloud revenue of $28.9 billion, a 91 percent jump, with TPU sales projected to grow from $9.4 billion to $138 billion by 2028.
- CoreWeave lost 11.7 percent of its stock value in one week, with long-term debt interest rates exceeding 16%, reflecting broader AI profitability struggles.
- CrowdStrike trades at 49.8x price-to-sales, far above the software industry average of 4.0x, partly due to its partnership with NVIDIA.
- The American Chamber of Commerce in Korea urged companies to train workers and manage risks when adopting AI tools like ChatGPT.
Anthropic files for massive IPO with $2 trillion valuation target
Anthropic, the maker of Claude AI, filed for an IPO that could value the company at up to $2 trillion. Revenue jumped to $16.3 billion in the first half of 2026 from about $1.2 billion a year earlier. The company reported a net loss of $80.7 billion and plans to list on Nasdaq under the ticker ANT. It warns that advanced AI could pose catastrophic risks to humanity.
Anthropic flags AI dangers while pursuing largest IPO ever
Anthropic's IPO filing devoted over 90 pages to risk factors including threats from advanced AI systems. The company said AI could pose catastrophic or existential risks if misused or left operating with high autonomy. Anthropic is seeking a valuation up to $2 trillion, which could surpass SpaceX's $1.77 trillion valuation. The company holds $131.7 billion in cash and generated $16.3 billion in revenue in the first half of 2026.
Anthropic files for biggest IPO ever as CEO warns of AI risks
Anthropic filed paperwork for what could be the largest IPO ever, potentially eclipsing SpaceX's $86 billion raise. The five-year-old company behind Claude reported revenue over $16 billion in the first half of the year, 14 times the prior year. Despite losses, it could be valued at $1.5 to $2 trillion. Founders could each gain as much as $28 billion if the company reaches the top expected value.
Anthropic warns of existential AI risks in IPO filing
Anthropic's IPO filing warns that its technology could help malicious actors develop dangerous weapons and that AI models could behave unpredictably. Revenue rose from $4.6 billion last year to over $33 billion in nine months. Five customers accounted for 25 percent of revenue. The company has signed computing and infrastructure commitments worth over $800 billion through 2045, including deals with Broadcom, SpaceX, Google, Amazon and Microsoft.
Tata Communications to invest $450 million in AI fibre network
Tata Communications plans to invest $400 to $450 million over three to five years to build a 10,000-km fibre network across India with 7.2 petabytes of capacity. The network will connect key data centres and AI hubs in three phases, starting with the Mumbai-Pune-Hyderabad-Chennai corridor. The company expects the data centre connectivity market in India could exceed $1 billion over the medium term. Revenue from the new network is expected to grow around FY28 to FY29.
Tata Communications plans $450M AI fiber network across India
Tata Communications will invest $400M to $450M over 3 to 5 years to build a long-haul fiber network called the Indian AI Super Highway. The project will cover about 10,000 km and carry 7.2 Petabytes of data across major Indian cities. Phase 1 connects Mumbai, Pune, Hyderabad and Chennai, with later phases reaching other hubs. Revenue is expected to begin in FY28-29, targeting a market worth over $1 billion.
AmCham calls for worker training as companies adopt AI
The American Chamber of Commerce in Korea held a workshop in Seoul to discuss AI adoption in businesses. Leaders said companies must train workers and manage risks to gain real productivity gains. Speakers included representatives from Google, OpenAI and other groups. They discussed tools like ChatGPT, workplace changes and legal risks for executives.
Neura Robotics raises $1.4 billion to build human-like AI robots
Neura Robotics, based in Metzingen, Germany, raised up to $1.4 billion in funding. The investment was led by Tether and backed by Nvidia and Amazon. The company is working on AI-powered robots that can perform complex human-like tasks. Its rise highlights Europe's push to compete in robotics and AI against China.
CrowdStrike stock seen as pricey amid growing AI security ties
CrowdStrike trades at a price-to-sales ratio of 49.8x, much higher than the software industry average of 4.0x and AI peers at about 15.5x. Its partnership with NVIDIA places it in the spotlight for AI security infrastructure. Some investors believe the stock is overvalued, while others see strong AI-driven growth ahead.
Anthropic IPO filing warns AI could threaten humanity
Anthropic's S-1 form was published by the SEC ahead of its expected initial public offering. The filing discusses risks tied to its AI technology, including the possibility of blackmail, sentience or harm to humans. The document also addresses the company's fast-growing revenues and high costs. It has sparked debate about the future of AI development.
AMD to buy World Labs for $8.2 billion to boost AI technology
AMD has agreed to acquire World Labs, an AI research lab led by Dr. Fei-Fei Li. The deal is valued at about $8.2 billion in stock and is expected to close by the end of 2026. World Labs specializes in spatial-intelligence models that create 3D environments from text, image, and video inputs. Dr. Fei-Fei Li will join AMD as executive vice president and chief scientist. The acquisition aims to strengthen AMD's AI hardware, software, and systems for an open ecosystem.
Microsoft security chief warns AI cannot be trusted
Microsoft Security EVP Hayete Gallot says frontier AI models cannot be trusted and AI agents have too much power in enterprises. She describes these agents as unpredictable agentic insiders that can collaborate to reach objectives. Microsoft recommends using Guardian Agents for real-time monitoring. The company's Project Perception system can be adapted to contain agentic insiders with deterministic controls.
AI industry struggles as companies fail to make a profit
Many AI startups are struggling to turn a profit despite heavy investor funding. CoreWeave, the largest neocloud, lost 11.7% of its stock value in one week after dropping over 25% the week before. Its long-term debt interest rate now exceeds 16%. The lack of revenue creates ripple effects across the broader economy and makes it harder for new startups to emerge. Experts suggest the industry needs to focus on practical applications.
Google Q3 results focus on AI growth and Meta Muse threat
Google is set to report Q3 results with analysts expecting earnings per share of $3.01 on revenue of $127.8 billion. The cloud unit is projected to see revenue jump 91% to $28.9 billion. Google may also earn significant revenue from selling custom Tensor Processing Units to third-party data centers. Analysts expect TPU sales to grow from $9.4 billion in late 2026 to $138 billion by 2028. Meta's Muse AI launch poses a potential long-term threat to Google's search business.
Sources
- Anthropic prepares for potential $2 trillion IPO valuation after reporting billions in revenue — company warns AI could pose ‘catastrophic existential risks to humanity’
- Anthropic warns of ‘catastrophic threat to humanity’ as it eyes $2 trillion valuation in biggest-ever IPO
- Artificial intelligence giant Anthropic files papers to sell...
- IPO-Bound Anthropic Warns Of 'Existential Risks To Humanity': What Investors Should Know
- Tata Communications To Invest $450 Mn In AI Fibre Network
- Tata Communications Announces USD 400 - 450M Investment to Build India
- AmCham stresses workforce training, risk management in AI transition
- In the A.I. ‘Robogym,’ Machines Train to Be More Like Humans
- CrowdStrike (CRWD) Stock May Look Fully Priced As AI Security Ties Expand
- Anthropic IPO warns its AI could blackmail people, become sentient, end humanity
- AMD to Acquire World Labs to Advance AI Compute and Infrastructure
- Microsoft Security chief says AI 'cannot be trusted' and AI agents are given too much power
- Artificial Intelligence Has A Big Problem: Nobody Wants To Pay
- Google to report Q3 results as investors focus on AI growth, weigh Meta Muse AI risk
Comments
Please log in to post a comment.