Chinese AI Companies Rush to Secure Nvidia Chips as U.S. Restrictions Fears Grow

Chinese artificial intelligence companies are rushing to secure Nvidia chips through overseas data centres amid fears of stricter U.S. export controls. Several companies are exploring ways to reserve computing capacity outside China to maintain access to advanced processors.

According to a Reuters report published on November 12, 2026, Chinese technology companies fear tighter U.S. restrictions. Washington could introduce stricter rules governing remote access to Nvidia chips already subject to export controls. These concerns have grown as U.S. authorities investigate suspected technology transfers to China and consider additional restrictions.

The development shows China’s continued reliance on advanced American semiconductors as domestic technology companies compete globally. Nvidia chips remain important for developing large AI models and expanding computing infrastructure.

Chinese AI Companies Seek Nvidia Chips Through Overseas Data Centres

Several Chinese AI companies are exploring overseas data centres to access advanced Nvidia chips amid growing US restrictions. Three people familiar with the discussions told Reuters that companies are considering reserving computing capacity before new restrictions take effect.

A major Chinese cloud provider has reportedly discussed securing Nvidia computing resources at overseas facilities. Other Chinese companies have also approached data centre operators to reserve access to advanced processors.

Smaller AI research labs are exploring similar options through cloud providers across Southeast Asia. These arrangements would allow Chinese developers to rent computing power without directly importing restricted Nvidia chips.

Overseas data centres offer Chinese companies an alternative way to access advanced American processors. These facilities provide computing resources for training and running large AI models, including generative AI applications.

ALSO READ: Nvidia-backed Nscale Left ByteDance Relationship Out of Main Filing as It Targets $35B IPO

Proposed U.S. Legislation Could Restrict Remote Access to Nvidia Chips

Chinese technology companies are also monitoring proposed U.S. legislation that could limit remote access to advanced American processors.

The proposed Remote Access Security Act (RASA) would expand the U.S. Commerce Department’s authority over restricted technology access. The U.S. House of Representatives passed the bill in January 2026, receiving 369 votes in favour and 22 against.

However, the legislation still needs Senate approval before it can become law. If approved, it could restrict Chinese companies from renting computing capacity powered by restricted Nvidia processors.

The proposed law could close an existing route for Chinese AI developers to access advanced American chips. This uncertainty has pushed companies to explore overseas data centre options before further restrictions take effect.

Nvidia CEO Jensen Huang Faces Scrutiny Amid China-Related Allegations

Nvidia CEO Jensen Huang Faces Scrutiny Amid China-Related Allegations

The growing concerns over Nvidia chips come amid reports linking CEO Jensen Huang to allegations involving suspected Chinese intelligence activities.

The New York Times reported that U.S. authorities identified Alice Huang as a suspected Chinese intelligence asset. She previously worked as an executive at Megaspeed, a Singapore-based cloud computing company.

Reuters said it could not independently verify the allegations. The Chinese government has also denied the accusations.

Megaspeed has faced scrutiny over claims that it helped Chinese customers access Nvidia chips restricted under U.S. export controls. However, the company has denied illegally diverting advanced processors to China.

The controversy has raised further questions about how Chinese companies obtain restricted American computing hardware.

Grey-Market Demand for Nvidia Chips Increases

Demand for Nvidia processors is also rising in China’s grey market as buyers prepare for tighter U.S. export controls. Traders and intermediaries are reportedly stockpiling advanced chips amid concerns about stricter enforcement and further restrictions.

A recent procurement case has highlighted the difficulties of enforcing U.S. semiconductor export restrictions. A Chinese university with reported ties to the People’s Liberation Army purchased Nvidia’s powerful H200 AI chips last month.

According to procurement records cited by Reuters, the university was already on the U.S. export blacklist.

Separately, U.S. prosecutors charged Super Micro Computer co-founder Yih-Shyan Wally Liaw and two others in March 2026. They were accused of conspiring to divert Nvidia-equipped servers to China through a Southeast Asian company.

According to prosecutors, the company purchased approximately $2.5 billion worth of Super Micro servers between 2024 and 2025. A substantial portion of those servers was allegedly redirected to customers in China.

ALSO READ: PLA-Linked Chinese University Bought Nvidia H200 AI Chips Despite US Export Ban

U.S. Export Controls Create Challenges for China’s AI Industry

Recent developments highlight the challenges facing Chinese AI companies as Washington tightens restrictions on advanced semiconductor exports. Nvidia processors remain essential for training large language models and expanding artificial intelligence computing infrastructure.

Chinese technology companies are exploring overseas data centres to maintain access to advanced Nvidia chips. However, proposed U.S. legislation could restrict these alternatives if Washington expands its semiconductor export control rules.

Rising demand for overseas Nvidia computing capacity also highlights the importance of advanced chips in global AI competition. Additional restrictions could put further pressure on Chinese developers as they expand their AI models and computing infrastructure.

For Nvidia, these developments create challenges in serving Chinese customers while complying with U.S. export regulations.

Posted in AI News | Tagged | Leave a comment

Google Faces EU Setback as Court Adviser Backs $1.7 Billion Antitrust Fine

Google has faced a fresh legal setback in Europe after a top EU court adviser backed regulators’ appeal. The case involves a €1.49 billion ($1.7 billion) antitrust fine previously overturned by a lower court. The original penalty concerned Google’s online advertising business and restrictions imposed on competing advertising providers.

On November 12, 2026, Advocate General Juliane Kokott recommended overturning the lower court’s 2024 judgment. She argued that the General Court made legal errors while examining Google’s advertising contracts and their effects on competition.

The dispute focuses on Google’s AdSense for Search service and its agreements with website publishers. Regulators alleged that these contracts restricted rival advertising providers from competing for website advertising space. The European Commission imposed the €1.49 billion fine in March 2019 over these practices.

According to the Court of Justice of the European Union’s official announcement, the lower court failed to properly assess Google’s contractual restrictions. Kokott believes the court did not adequately examine how these practices affected competition and excluded rival advertising providers.

EU Court Adviser Challenges Decision to Overturn Google’s €1.49 Billion Fine

The European Commission initially fined Google €1.49 billion in March 2019 over alleged violations of EU competition rules. Its investigation found that Google imposed restrictive conditions in agreements with website publishers between 2006 and 2016.

These restrictions affected publishers using Google’s AdSense for Search advertising service. The agreements limited publishers’ ability to display advertisements from competing search advertising providers.

The Commission identified three main types of restrictive conditions in Google’s publisher agreements. These included exclusivity requirements, advertising placement restrictions and approval requirements for competing advertisements.

According to the Commission, these conditions reduced opportunities for rival advertising providers to compete for website advertising space. Regulators concluded that Google’s contractual practices violated European Union antitrust rules.

Google stopped using or changed the disputed contractual conditions in September 2016. However, the company challenged the Commission’s findings and appealed against the €1.49 billion penalty.

In September 2024, the EU General Court overturned the European Commission’s decision to fine Google. The court found that the Commission had failed to properly assess important details, particularly the duration of the contractual restrictions.

Following this ruling, the European Commission appealed to the Court of Justice of the European Union. It challenged the General Court’s decision to cancel Google’s antitrust penalty.

Advocate General Says EU Court Made Legal Errors in Google’s Antitrust Case

Advocate General Says EU Court Made Legal Errors in Google's Antitrust Case

In her latest opinion, Advocate General Juliane Kokott said the General Court applied excessive legal requirements. These requirements concerned the duration and competitive effects of Google’s contractual restrictions on website publishers.

She also argued that the lower court failed to connect its findings with relevant EU competition laws. In particular, its assessment did not properly address the legal standards for identifying abuse of market dominance.

Kokott recommended that the Court of Justice overturn the General Court’s September 2024 ruling. She also suggested returning the case to the lower court for another examination. However, the Advocate General’s opinion does not legally bind the Court of Justice. 

The final ruling will come from the European Union’s highest court, which is expected to decide the case in 2027. The Court of Justice often considers its Advocates General’s recommendations when reviewing complex legal disputes. However, judges can reach a different conclusion because these opinions are not legally binding.

ALSO READ: Alphabet Loses Nearly $500 Billion After Google Unveils Gemini 4 Carbon

Google Faces Continued EU Antitrust Pressure Over Its Business Practices

The Google AdSense case forms part of the company’s wider antitrust disputes with European competition regulators. Google has faced several major penalties over allegations involving its search engine, Android operating system and advertising technology business.

The AdSense penalty was one of six major antitrust fines imposed on Google by the European Commission. Together, these penalties amounted to approximately €10.38 billion over nearly two decades.

In July 2026, the Court of Justice upheld Google’s reduced €4.125 billion fine over its Android operating system practices. The European Commission had found that Google used contractual restrictions to strengthen its search engine and Chrome browser on Android devices.

Google also faces separate regulatory challenges over its advertising technology business and compliance with the EU’s Digital Markets Act. These cases add to growing regulatory pressure on the company’s business practices across Europe.

The latest Advocate General’s opinion marks another development in Google’s ongoing European antitrust disputes. If the Court of Justice follows Kokott’s recommendation, the General Court must reconsider the original AdSense penalty.

However, Google has not automatically lost its previous court victory, and the €1.49 billion fine remains overturned. The final outcome depends on the Court of Justice’s upcoming judgment and any further proceedings before the lower court.

Posted in AI News | Tagged | Leave a comment

How Many Subscribers Does Netflix Have Worldwide 2026

Netflix is one of the world’s leading streaming platforms, with millions of people watching its movies, TV shows, documentaries, and original series. Netflix has approximately 350 million subscribers worldwide in 2026, making it one of the largest streaming services globally. 

Its subscriber base has grown steadily over the years, rising from 100 million in 2017 to 325 million in 2025. The company continues to attract viewers through its original content, international shows, and growing library of movies and series. 

This article covers the latest Netflix subscriber statistics for 2026, including its global subscriber growth, country-wise figures, age and gender breakdown, annual revenue, and viewing trends.

Key Statistics Worldwide (2026)

  • Netflix is estimated to have approximately 350 million subscribers worldwide in 2026.
  • Netflix’s global subscriber base reached 325 million in 2025, up from 302 million in 2024.
  • US households spent an average of 3.5 hours watching Netflix daily in June 2026.
  • The United States has the largest Netflix subscriber base, with an estimated 87.75 million members in 2026.
  • The United Kingdom ranks second, with approximately 19.83 million Netflix subscribers.
  • India has an estimated 13.32 million Netflix subscribers in 2026.
  • Europe, the Middle East, and Africa (EMEA) became Netflix’s largest subscriber region in 2024, with 101.13 million paying subscribers.
  • Around 72% of US adults aged 18 to 29 have a Netflix subscription, the highest rate among the listed age groups.
  • War Machine was Netflix’s most-watched title in the first half of 2026, attracting 147 million views.
  • Netflix recorded 97 billion viewing hours in the first half of 2026, compared to 95 billion during the same period in 2025.

How Many Subscribers Does Netflix Have Worldwide?

Netflix has approximately 350 million subscribers worldwide in 2026, based on estimates. Netflix is one of the most popular streaming services globally, with millions of people subscribing to watch movies, TV shows, documentaries, and original content. 

The platform has expanded its global subscriber base over the years, growing from just 400,000 subscribers in 2001 to 100 million in 2017. Netflix crossed 200 million subscribers in 2020 and continued to attract new users through its growing content library and international expansion.

Its subscriber count reached 302 million in 2024 and 325 million in 2025. In 2026, Netflix is expected to reach 350 million subscribers, reflecting continued growth in its worldwide streaming audience.

How Many Subscribers Does Netflix Have Worldwide?
YearNetflix SubscribersYoY Growth
20010.4 million–
20020.8 million100.00%
20031.41 million76.25%
20042.48 million75.89%
20054.02 million62.10%
20066.15 million52.99%
20077.32 million19.02%
20089.16 million25.14%
200911.89 million29.80%
201018.26 million53.57%
201124.3 million33.08%
201230.36 million24.94%
201341.43 million36.46%
201454.47 million31.48%
201570.83 million30.03%
201689.09 million25.78%
2017110.64 million24.19%
2018139.25 million25.86%
2019167.09 million19.99%
2020203.66 million21.89%
2021221.84 million8.93%
2022230.75 million4.02%
2023260.28 million12.80%
2024302 million16.03%
2025325 million7.62%
2026 (Est.)350 million7.69%

Countries With the Most Netflix Subscribers

Netflix is available in more than 190 countries and has built a large subscriber base across international markets. However, the number of Netflix subscribers varies widely depending on market size, internet access, and streaming demand.

The United States leads the Netflix subscriber rankings with an estimated 87.75 million members in 2026. This figure is more than four times the United Kingdom’s estimated subscriber count of 19.83 million. 

Brazil and Germany follow with approximately 17.87 million subscribers each, making them major Netflix markets. Mexico and France also rank among the countries with the largest Netflix subscriber bases.

Countries With the Most Netflix Subscribers
RankCountryEstimated Netflix Subscribers (2026)
1United States87.75 million
2United Kingdom19.83 million
3Brazil17.87 million
4Germany17.87 million
5Mexico14.95 million
6France14.62 million
7India13.32 million
8Japan9.75 million
9Canada9.75 million
10South Korea9.00 million

India and Japan have fewer Netflix subscribers but remain important markets for the company’s international growth. India recorded its highest-ever quarterly paid subscriber additions during the third quarter of 2026. Japan also contributed significantly to Netflix’s membership growth during the first quarter of 2026.

Netflix Subscribers by Region

Netflix has a large subscriber base across North America, Europe, Latin America, and Asia-Pacific. The number of paying subscribers has increased across all major regions between 2022 and 2024. However, Europe, the Middle East, and Africa (EMEA) became Netflix’s largest subscriber region by 2024.

In 2022, the United States and Canada (UCAN) had approximately 74.58 million paying subscribers. EMEA followed closely with 73.73 million subscribers, making both regions Netflix’s largest markets. By 2024, EMEA had moved ahead of North America with more than 101 million paying subscribers.

RegionPaying Subscribers (2022)Paying Subscribers (2024)
UCAN (United States and Canada)74.58 million89.63 million
EMEA (Europe, Middle East, and Africa)73.73 million101.13 million
LATAM (Latin America)39.61 million53.33 million
APAC (Asia-Pacific)33.72 million57.54 million

Netflix Subscribers by Age Group

Netflix subscription rates vary across age groups, with younger and middle-aged adults showing the highest adoption. In the United States, 72% of adults aged 18 to 29 have a Netflix subscription, making them the largest age group by subscription rate. 

Netflix adoption remains strong among adults aged 30 to 49 at 70% and those aged 50 to 64 at 69%. However, the subscription rate drops to 44% among adults aged 65 and older.

Age GroupShare of People With a Netflix Subscription
18 to 29 years72%
30 to 49 years70%
50 to 64 years69%
65 years and older44%

Netflix Subscribers by Gender

Netflix’s global audience is almost evenly divided between male and female users. Women account for 51% of Netflix users, while men represent the remaining 49%. This small difference shows that Netflix attracts viewers from both genders across its global audience. 

Its wide range of content, including drama, action, comedy, documentaries, and reality shows, appeals to different viewing preferences.

GenderShare of Netflix Users
Male49%
Female51%

Netflix Annual Revenue Growth

Netflix generated $45.1 billion in annual revenue in 2025, compared to $39 billion in 2024. This represents a 15.8% increase in revenue year over year. The company’s revenue has grown steadily over the past decade, rising from $3.1 billion in 2011 to $45.1 billion in 2025. 

Netflix earns most of its revenue through monthly subscriptions across its global streaming platform. Its growing subscriber base, subscription price increases, and expansion into international markets have supported this growth. 

The company has also introduced advertising-supported plans and expanded its content library to attract more viewers. The following table shows Netflix’s annual revenue growth from 2011 to 2025.

YearNetflix Revenue (USD Billion)
2011$3.1 billion
2012$3.5 billion
2013$4.3 billion
2014$5.4 billion
2015$6.7 billion
2016$8.8 billion
2017$11.6 billion
2018$15.7 billion
2019$20.1 billion
2020$24.9 billion
2021$29.6 billion
2022$31.6 billion
2023$33.7 billion
2024$39 billion
2025$45.1 billion

Source: Businessofapps

How Much Time Do People Spend Watching Netflix?

U.S. households spent an average of 3.1 hours watching Netflix daily in March 2026. By June 2026, average daily viewing time had increased to 3.5 hours, marking a 24-minute rise. Netflix recorded higher daily viewing time than other major streaming platforms, including Amazon Prime Video, Disney+, Hulu, and HBO Max.

Streaming SitesMarch 2026June 2026
Netflix3.1 hours3.5 hours
Disney+2.1 hours2.2 hours
HBO Max1 hour2.1 hours
Apple TV1.4 hours2.6 hours
Amazon Prime Video1.4 hours2.6 hours
Hulu2.8 hours2.7 hours
Paramount+3.1 hours3.5 hours
Peacock TV2.2 hours2.2 hours

Source: Emarketer

Netflix Viewing Hours in 2026

Netflix recorded 97 billion viewing hours during the first half of 2026, setting a new record for its global streaming platform. This marked an increase from 95 billion hours in the first half of 2025 and 96 billion hours in the second half of 2025. The growth shows that Netflix continues to attract viewers across different countries and content categories.

PeriodTotal Netflix Viewing Hours
First Half of 202595 billion
Second Half of 202596 billion
First Half of 202697 billion

Most-Watched Netflix Titles in the First Half of 2026

Netflix’s most-watched titles in the first half of 2026 included several popular movies and TV shows. War Machine topped the list with 147 million views, followed by The Rip with 136 million views. 

Swapped and Apex also performed well, with 131 million and 129 million views, respectively. Among TV shows, Bridgerton Season 4 received 100 million views, while ONE PIECE reached 69 million views. Other popular titles included His & Hers, Thrash, and The Night Agent.

Netflix TitleViews
War Machine147 million
The Rip136 million
Swapped131 million
Apex129 million
His & Hers104 million
Thrash100 million
Bridgerton Season 4100 million
I Will Find You64 million
The Crash65 million
ONE PIECE69 million
The Night Agent63 million
The Lincoln Lawyer59 million
Office Romance58 million
Run Away50 million
Teach You a Lesson48 million

Source: Netflix

Wrapping Up

Netflix continues to hold a strong position in the global streaming market, with an estimated 350 million subscribers worldwide in 2026. Its subscriber base, annual revenue, and total viewing hours have grown significantly over the years. 

The United States remains its largest market, while countries such as the United Kingdom, Brazil, Germany, and India contribute to its international reach. Netflix also attracts viewers across different age groups and offers content in multiple languages to reach a wider audience. Its growing content library, international expansion, and original productions continue to support its position in the streaming industry.

Posted in Statistics | Leave a comment

Siemens Energy Secures €15 Billion in AI Data Center Orders, Raises 2027 Profit Forecast

German energy technology company Siemens Energy has reported around €15 billion in orders from AI data-center projects as demand for electricity infrastructure continues to grow. The company has also raised its fiscal 2027 profit margin forecast, supported by strong demand for gas turbines and power grid equipment.

The announcement came on November 11, 2026, alongside the company’s record financial results for fiscal 2026. Siemens Energy reported a total order backlog of €164 billion, with AI data-center orders accounting for around 9.1% of the total.

The rapid expansion of AI data centers is increasing demand for electricity generation and transmission equipment. These facilities need large amounts of electricity to run AI servers, train large language models, and handle other computing tasks.

Siemens Energy expects rising AI investments to create more demand for gas-fired power plants and electricity grid infrastructure. North America is expected to remain a major market for this growth.

Siemens Energy Reports €15 Billion in Orders From AI Data Centers

Siemens Energy has received around €15 billion in orders linked to AI data centers. CEO Christian Bruch said these orders include gas turbines and power grid equipment needed to supply electricity to large data centers.

The orders account for approximately 9.1% of the company’s record €164 billion backlog. This backlog covers orders from all its business divisions and gives the company a clearer view of future revenue.

Bruch said demand for data-center projects remains strong, with no major changes in customer activity. However, he pointed out that some earlier forecasts for data-center expansion may have been too optimistic.

Despite this, rising demand for AI computing continues to create opportunities for Siemens Energy. As technology companies expand their AI infrastructure, they need more electricity generation capacity and reliable power grids.

Siemens Energy expects data-center construction to support further investment in power generation and grid equipment. North America is likely to remain a key market as demand for AI infrastructure continues to grow.

ALSO READ: AI Data Center Energy Statistics

Siemens Energy Increases 2027 Revenue Growth and Profit Forecast

Siemens Energy has increased its financial outlook for fiscal 2027 as demand for its power generation and grid equipment continues to grow.

The company now expects its profit margin before special items to reach 14% to 16% in fiscal 2027. This is higher than the 12.4% margin reported for fiscal 2026.

Siemens Energy also forecasts comparable revenue growth of 14% to 16% for fiscal 2027. It expects net income to reach approximately €5 billion, while pre-tax free cash flow could rise to around €10 billion.

The company has also set new financial targets for fiscal 2030. It aims to achieve annual comparable revenue growth of 12% to 14% and a profit margin before special items of 20% to 22%.

These targets reflect Siemens Energy’s expectations of continued growth in electricity demand. Rising investments in power generation, electricity grids, and AI data-center infrastructure are expected to support its business expansion.

Siemens Energy Reports Record €164 Billion Order Backlog in 2026

Siemens Energy Reports Record €164 Billion Order Backlog in 2026

Siemens Energy delivered strong financial results for fiscal 2026, which ended on September 30, 2026. The company reported higher revenue, rising orders, and a record order backlog of €164 billion.

Its comparable revenue grew 15.5% year over year to €44.2 billion. Total orders also increased by 18.8% to €68.1 billion. This growth pushed the company’s order backlog to a record €164 billion.

Profit before special items more than doubled to approximately €5.49 billion. Net income also rose to €4.13 billion, compared with €1.69 billion in fiscal 2025. Pre-tax free cash flow nearly doubled as well. It reached €8.8 billion, up from €4.66 billion in the previous fiscal year.

Siemens Energy said it met or exceeded all its previously revised financial targets. Its wind energy subsidiary, Siemens Gamesa, also reached break-even in profit before special items. This marks an improvement after several years of financial losses and restructuring efforts.

ALSO READ: Nvidia and Broadcom Face Less Risk From US AI Data Center Power Shortage

Siemens Energy Announces €9 Billion Share Buyback Through 2030

Siemens Energy has announced plans to launch an additional share buyback programme worth up to €9 billion. The programme is expected to continue through fiscal 2030. The company has also proposed a dividend of €2.30 per share for fiscal 2026. This is more than three times the €0.70 per share paid to shareholders in the previous fiscal year.

These announcements reflect Siemens Energy’s stronger financial position and its expectations for continued growth in the energy infrastructure market.

The company’s latest results also show how rising AI investments are creating new demand for energy equipment. As technology companies expand their AI data centers, they need more gas turbines, transformers, and electricity grid equipment to support their operations.

This growing demand could create new business opportunities for Siemens Energy and other energy equipment suppliers as global investment in AI infrastructure increases.

Posted in AI News | Leave a comment

PLA-Linked Chinese University Bought Nvidia H200 AI Chips Despite US Export Ban

A Chinese university with links to the People’s Liberation Army (PLA) has purchased two Nvidia H200 AI chips despite being on the US government’s export blacklist. The purchase raises fresh concerns about how Chinese institutions continue to access advanced Nvidia processors despite US export restrictions.

According to procurement records reviewed by Reuters, Northwestern Polytechnical University bought two Nvidia H200 chips for 874,700 yuan (approximately $123,000) in October 2026. The university purchased the processors through a Chinese technology supplier instead of buying them directly from Nvidia.

The purchase shows potential gaps in US restrictions on AI chip exports to China. It also raises questions about how domestic suppliers and computing service providers help Chinese institutions access restricted AI hardware.

Chinese University Buys Two Nvidia H200 AI Chips for $123,000 Despite US Restrictions

Northwestern Polytechnical University in China purchased two Nvidia H200 AI chips for 874,700 yuan (approximately $123,000), according to a procurement notice published on its official website on October 21, 2026.

The purchase was made for the university’s management school. Chinese technology supplier Shaanxi Bosuan Network Technology handled the transaction. The Nvidia H200 is one of the company’s most powerful AI processors. It is built to handle demanding computing tasks, including training and running advanced artificial intelligence models. 

The chip is also used for scientific research, large-scale AI development, and other high-performance computing workloads. The purchase has drawn attention because Northwestern Polytechnical University has been under US export restrictions for more than two decades. These restrictions limit its access to certain American technologies.

The university is also a member of China’s Seven Sons of National Defense, a group of seven universities with close ties to the country’s military and defence research programmes. Its research areas include military aircraft, missile systems, and satellite technology.

US Export Restrictions on Northwestern Polytechnical University Explained

US Export Restrictions on Northwestern Polytechnical University Explained

The US Department of Commerce added Northwestern Polytechnical University to its Entity List in 2001 over concerns about its alleged involvement in missile proliferation and military research.

The restrictions became tighter in 2019. Since then, US companies have generally faced a policy of licence denial when seeking to export controlled technologies to the university.

The US Entity List limits access to certain American products and technologies without government approval. These restrictions also cover advanced computing equipment, including high-performance AI chips and processors.

In January 2026, the Trump administration began allowing some Nvidia H200 chip exports to China on a case-by-case basis. However, this policy change did not remove export restrictions on blacklisted organisations such as Northwestern Polytechnical University.

The university’s latest purchase has therefore raised questions about how it obtained the Nvidia H200 processors despite existing US export restrictions.

Chinese University Rented Servers With Restricted Nvidia H200 AI Chips

The purchase of two Nvidia H200 processors was not the only way Northwestern Polytechnical University accessed advanced AI computing hardware. Procurement records show that the university also rented computing resources from Chinese technology companies through separate contracts.

In January 2026, the university’s cybersecurity school awarded two contracts for access to servers equipped with eight Nvidia H200 chips each. Both contracts covered a period of seven months.

The contracts were worth 441,700 yuan and 441,000 yuan, respectively. The university planned to use these computing resources for AI model fine-tuning and reinforcement learning. The university also rented servers equipped with eight Nvidia A100 chips and eight A800 chips.

Both Nvidia A100 and A800 processors face US export restrictions on sales to Chinese entities. These arrangements highlight a challenge for Washington’s efforts to limit China’s access to advanced AI chips. Chinese universities and research institutions can rent computing power from domestic cloud providers instead of buying processors directly from US chipmakers.

Multiple Chinese Universities Acquired Nvidia AI Chips Despite US Export Restrictions

The procurement records suggest that access to restricted Nvidia AI chips extends beyond Northwestern Polytechnical University. Several other Chinese universities and research institutions have also purchased these processors or rented computing services that use them.

Reuters reviewed 26 Chinese procurement and corporate records from 2026. The documents identified multiple institutions that had obtained restricted Nvidia processors through direct purchases or computing service contracts. Other reported purchases include:

  • Shenzhen University: Purchased three Nvidia H100 processors for 769,800 yuan in September 2026.
  • Wuhan University of Technology: Purchased a server containing an Nvidia H100 chip for 355,000 yuan in June. It also bought four Nvidia A100 processors for 344,000 yuan.
  • Beihang University: Purchased four Nvidia A800 processors for 489,000 yuan in April. The university is also a member of China’s Seven Sons of National Defense.
  • University of Chinese Academy of Sciences: Procurement notices published in July and September listed suppliers offering computing services powered by Nvidia H800 and A800 processors.

Several of these universities have links to China’s defence sector and military research programmes.

The records also show that some institutions accessed restricted Nvidia processors through Chinese technology suppliers. This allowed them to obtain computing resources without purchasing the chips directly from Nvidia.

However, the procurement documents did not provide clear details about the chips’ original sources or the physical locations of the rented servers. As a result, it remains unclear how these institutions obtained the restricted Nvidia processors and where the hardware was being operated.

ALSO READ: Nvidia GPU Black Market Statistics 2026

Nvidia Faces US Lawmakers’ Scrutiny Over AI Chip Sales to China

Nvidia and CEO Jensen Huang are facing growing scrutiny from US lawmakers over the company’s AI chip exports to China. Huang has argued that allowing Nvidia to sell advanced AI chips to China would help the US maintain its technological lead. 

He has also warned that export restrictions could push Chinese companies towards domestic chipmakers. However, US lawmakers have raised national security concerns. They argue that advanced AI processors could help China develop military AI systems, improve intelligence analysis, and advance weapons research.

Democratic and Republican lawmakers have called for stricter export controls. Their proposals include chip-tracking requirements and greater congressional oversight of AI chip sales. Three Republican committee chairmen have also requested a briefing with Huang on November 18. 

The request concerns his reported contacts with an individual allegedly linked to Chinese intelligence. Nvidia, Northwestern Polytechnical University, and the US Commerce Department’s Bureau of Industry and Security did not respond to Reuters’ requests for comment on the reported purchases.

ALSO READ: Nvidia CEO Jensen Huang Reportedly Targeted by Suspected Chinese Spy, US Intelligence Says

US Export Controls Face Challenges as China Expands Access to Advanced AI Chips

US Export Controls Face Challenges as China Expands Access to Advanced AI Chips

The latest data raise questions about the effectiveness of US export restrictions on advanced AI chips to China.

Although Washington has restricted direct chip sales, Chinese technology suppliers and cloud service providers appear to offer other ways for universities and research institutions to access restricted Nvidia processors.

Renting computing power also makes export enforcement more difficult. US authorities may struggle to track where restricted chips are located, who operates them, and which organisations use their computing resources.

The US government must therefore look beyond direct chip shipments. It also needs to monitor domestic suppliers, cloud computing services, and other channels that could provide restricted AI hardware to Chinese institutions linked to military research.

The latest procurement records could increase pressure on Washington to strengthen export controls as the US and China compete for leadership in advanced AI technology.

Posted in AI News | Tagged | Leave a comment

AI Cash-Burn Fears Trigger Nearly $50 Billion Foreign Outflow From Asian Stocks

Foreign investors pulled nearly $50 billion from Asian stock markets in October 2026 as concerns over artificial intelligence spending increased. Rising cash burn at major technology companies and possible regulatory restrictions also weakened investor confidence across the region.

The sell-off hit technology-heavy markets, particularly Taiwan and South Korea, where investors reduced their holdings in semiconductor and AI-related stocks. Data from the London Stock Exchange Group (LSEG) showed net foreign equity outflows of $49.77 billion across seven Asian markets in October. 

This marked the region’s largest monthly foreign investment withdrawal since June 2026, highlighting growing concerns about AI infrastructure spending. Investors are increasingly questioning whether major technology companies can generate enough revenue to justify their massive AI infrastructure investments.

Foreign selling has also continued in November, with another $22.6 billion leaving these Asian equity markets so far. A Reuters report published on November 11, 2026, shows the continued withdrawals amid growing concerns over AI spending and investment returns.

Taiwan and South Korea Lead Asia’s Foreign Stock Market Outflows

Taiwan and South Korea recorded Asia’s largest foreign equity outflows in October as investors reduced their stock market holdings. Foreign investors withdrew a net $30.68 billion from Taiwanese stocks during October, marking a record monthly outflow.

This was Taiwan’s largest foreign equity withdrawal since January 2008, highlighting growing concerns among overseas investors. South Korea also experienced heavy selling, with foreign investors pulling approximately $14.59 billion from its equity market.

The withdrawals reflect growing investor concerns about technology stocks, particularly companies linked to semiconductors and artificial intelligence infrastructure. Foreign selling has continued into November, adding further pressure on both Asian equity markets.

Taiwan has recorded another $19.76 billion in foreign equity outflows so far this month, while South Korea lost approximately $8.32 billion. Both countries have strong links to the global semiconductor supply chain and manufacture hardware used in AI data centres.

However, uncertainty surrounding future AI infrastructure spending has raised concerns about technology valuations and expected corporate earnings.

ALSO READ: AI Stocks Drop After Tech Leaders Call for Slower AI Development

Investors Question Returns as Big Tech Increases AI Infrastructure Spending

Investors Question Returns as Big Tech Increases AI Infrastructure Spending

Growing doubts about the profitability of artificial intelligence investments have emerged as a major reason behind the recent stock market selloff. Major technology companies, especially large cloud providers, are spending billions of dollars to expand their AI infrastructure.

Their investments cover advanced processors, servers, data centres, electricity systems, and networking equipment needed to support AI services. However, investors are questioning whether future revenue from AI services will cover these rising infrastructure costs.

Increasing capital expenditure and cash burn have also raised concerns about the financial risks of expanding AI operations. Wael Makarem, lead financial market strategist at brokerage firm Exness, linked the withdrawals to growing concerns about AI infrastructure spending and profitability.

He said investors were reducing their exposure to AI-related investments as doubts about future returns continued to increase. These concerns have also affected Asian stock markets, where major semiconductor and technology companies supply hardware for AI infrastructure.

ALSO READ: AI Company Burn Rate Statistics 2025-2026

US AI Regulations Raise Concerns Over Data Centre Expansion

Possible changes to US AI regulations have created additional uncertainty for investors concerned about future technology investments. Ahead of the US midterm elections, investors worried that stricter environmental reviews could delay new data centre projects.

Additional AI regulations could also slow data centre construction and reduce demand for advanced semiconductors and related equipment. Recent Democratic victories in several states have kept these concerns alive, particularly where opposition to new data centres continues growing.

Investors are now assessing whether tighter environmental rules and regulatory oversight could affect major technology companies’ expansion plans. These concerns have added pressure on Asian technology stocks, especially markets heavily dependent on semiconductor exports and AI infrastructure demand.

Asian Stock Markets Underperform Global Equities Amid Foreign Investor Sell-Off

Foreign investor withdrawals came as Asian stock markets recorded steeper losses than global equities during October 2026. The MSCI Asia Pacific ex-Japan index dropped 6.91% in October, significantly underperforming broader international stock market benchmarks.

By comparison, the MSCI All Country World Index declined just 1.69% during the same period. The wider gap indicates that investors reduced their exposure to Asian equities more sharply than global stocks.

Scarlett Liu, Asia-Pacific equity derivatives strategist at BNP Paribas, said international investors may have favoured US stocks over Asian markets. The relative strength of US equities may have encouraged investors to move their money away from Asian stock markets.

Rising US Treasury yields may have also influenced investment decisions, making government bonds more attractive to investors. The yield on 10-year US Treasury securities reached 5.3% in early October, increasing the appeal of these investments.

Foreign Investors Return to Indian Stocks With $5.68 Billion in November Inflows

India experienced foreign investor withdrawals for the second consecutive month in October, adding pressure to its equity market. Foreign investors withdrew a net $3.6 billion from Indian stocks during October, raising concerns about overseas investment trends.

However, foreign investment flows have reversed in November, with Indian equities attracting approximately $5.68 billion in net inflows. HSBC has also upgraded its outlook for Indian equities to overweight, citing stronger corporate earnings and more reasonable valuations.

Herald van der Linde, HSBC’s head of equity strategy for Asia Pacific, said Indian stocks offer better investment opportunities. He pointed to the country’s earnings outlook and stock valuations as key reasons behind the improved market assessment.

Meanwhile, several other Asian markets recorded comparatively smaller changes in foreign investment flows during October.

Asian MarketOctober 2026 Foreign Investment
Taiwan$30.68 billion
South Korea$14.59 billion
India$3.6 billion
Indonesia$938 million
Thailand$199 million
Vietnam$122 million
Philippines$109 million

Source: Reuters

AI Investment Concerns Put Asian Technology Stocks Under Pressure

AI Investment Concerns Put Asian Technology Stocks Under Pressure

Nearly $50 billion in foreign equity outflows highlight growing concerns about the rising costs of global artificial intelligence investments. Although demand for AI chips, computing systems, and data centre capacity remains strong, investors are questioning future financial returns.

Technology companies face increasing pressure to prove that their AI investments can generate enough revenue to cover rising costs. Possible changes to US AI regulations have added uncertainty, particularly for Asian economies reliant on semiconductor production and technology exports.

Another $22.6 billion has left Asian equity markets in November so far, adding pressure on regional stock markets. Investors will closely watch AI infrastructure spending, corporate earnings, US regulations, and broader market conditions before making investment decisions.

The recent foreign investor withdrawals do not necessarily signal weaker long-term demand for AI technologies and infrastructure. However, the sell-off suggests investors are becoming more cautious about companies exposed to the global AI boom.

Posted in AI News | Leave a comment

Alibaba Launches Qwen4-Max Preview With 1 Million-Token Context Window

Alibaba has launched a preview of Qwen4-Max, its next-generation flagship AI model, with a context window of up to 1 million tokens. The model can process text, images, and videos. It is designed to handle complex coding tasks, long documents, research, and AI agent workflows.

Announced on November 11, 2026, Qwen4-Max is the first preview release in Alibaba’s upcoming Qwen4 model family. The company is offering early access through its Token Plan subscription and several AI platforms, including the Qwen app and Qoder.

The launch adds a new flagship model to Alibaba’s AI lineup as competition grows among companies developing advanced AI models. Major areas of competition include longer context windows, better reasoning, and stronger coding performance.

Alibaba Qwen4-Max Supports 1 Million-Token Context Window

Alibaba’s Qwen4-Max AI model supports a context window of up to 1 million tokens, allowing it to process large volumes of information in a single conversation. This makes the model suitable for tasks that require analyzing lengthy documents, large codebases, and multiple sources of information.

The larger context window could help developers review entire software projects, summarize long reports, and complete complex tasks without splitting information into smaller sections. Qwen4-Max also supports image and video inputs, allowing users to analyze visual content alongside text. This expands its capabilities beyond text-based AI interactions.

According to Alibaba Cloud’s documentation, Qwen4-Max includes deep reasoning, function calling, web search, and structured output capabilities. The model also supports OpenAI’s Chat Completions and Responses APIs, along with Anthropic’s Messages API. This allows developers to integrate Qwen4-Max into existing applications with fewer compatibility changes.

ALSO READ: Alibaba AI Models Compared (2026): Qwen 2.5 & More

Alibaba Qwen4-Max Preview Availability, Access, and API Details

Alibaba Qwen4-Max Preview Availability, Access, and API Details

Alibaba has released Qwen4-Max Preview through its Token Plan subscription. The preview is available from November 11 to November 25, 2026, giving users a limited period to test the new AI model.

Developers can access Qwen4-Max through the Qwen app, Qoder, QwenWork, and QwenStudio. However, API access is restricted to Token Plan subscribers during the preview. Alibaba has not enabled pay-as-you-go API access for this release.

The model is available through the API under the identifier qwen4-max-preview. It supports a maximum output of 393,216 tokens, covering both reasoning tokens and the final response.

Alibaba has confirmed that the preview model will remain unchanged during the testing period. The company plans to discontinue access to this version after November 25, 2026.

Qwen4-Max Brings New Capabilities for Coding and AI Agent Workflows

Alibaba is positioning Qwen4-Max as an AI model for coding, software development, and multi-step tasks. The model is aimed at professional users who need advanced reasoning and the ability to handle complex instructions. According to Alibaba, Qwen4-Max can support several tasks, including:

  • Software development: Writing code and handling complex programming tasks.
  • Frontend development: Creating websites and designing user interfaces.
  • Office productivity: Processing documents and completing workplace tasks.
  • AI agent workflows: Using external tools to perform tasks across different applications.
  • Research and data analysis: Reviewing large volumes of information and handling long-context tasks.

These features could make Qwen4-Max useful for developers building AI agents and software applications. Its 1 million-token context window could also help users work with large codebases, lengthy documents, and complex datasets.

However, Alibaba has not yet published independent benchmark results for Qwen4-Max. Its claims about coding, reasoning, and AI agent performance have not been independently verified.

ALSO READ: China May Keep Its Best AI Models at Home and Sell Access Instead

Qwen4-Max Preview Signals the Next Phase of Qwen AI Models

Qwen4-Max Preview Signals the Next Phase of Qwen AI Models

The launch of Qwen4-Max Preview marks an important step in Alibaba’s next generation of Qwen AI models. Its 1 million-token context window and support for image and video inputs put it in competition with other advanced AI models focused on coding, research, and AI agent development.

The preview release also gives developers a chance to test Alibaba’s latest AI model and explore its capabilities before a wider launch.

However, Alibaba has not announced a general release date or final API pricing for Qwen4-Max. The company has also not confirmed whether it plans to release the model’s open weights. A complete technical report and independent benchmark results are also unavailable.

For now, developers will need to test Qwen4-Max to assess its coding, reasoning, and AI agent capabilities. These results will help determine how it compares with competing AI models from other companies.

Posted in AI News | Leave a comment

Meta Launches Muse Ember AI Model at $0.10 Per Million Tokens

Meta has launched Muse Ember, a low-cost AI model priced at $0.10 per million input tokens. The new Meta AI model targets developers and businesses looking for affordable solutions to run AI agents. Its launch also adds to the competition with OpenAI and Anthropic in the AI model market.

Announced on November 9, 2026, Muse Ember is built to handle repetitive tasks in AI agent workflows. These include coding subtasks, data extraction, document processing, and tool calls. Meta aims to help developers manage large volumes of AI tasks while keeping inference costs low.

Muse Ember is part of Meta’s Muse AI model family. Developers can access it through the Meta Model API and Muse Code. The model also offers a context window of up to 1,048,576 tokens and accepts text, images, audio, and video as inputs.

With Muse Ember, Meta is targeting businesses that need affordable AI models for large-scale automation. The company says the model delivers competitive benchmark results while costing less to run than several larger AI models.

Meta Sets Muse Ember AI Pricing at $0.10 Per Million Input Tokens

Meta has introduced competitive pricing for its Muse Ember AI model. The company charges $0.10 per million input tokens, $0.50 per million output tokens, and $0.004 per million cached input tokens.

These rates apply to requests with up to 200,000 input tokens. For longer requests, Meta charges $0.30 per million input tokens and $1.50 per million output tokens. The cached input rate increases to $0.012 per million tokens.

The low pricing makes Meta Muse Ember an affordable option for developers handling large volumes of AI requests. It could also help businesses reduce the cost of running AI agents that perform repetitive tasks.

According to Meta, Muse Ember costs around $0.14 per task in Artificial Analysis evaluations. Its larger Muse Spark 1.3 model costs approximately $2.39 per task.

Meta is targeting developers and businesses that need AI agents to handle multiple small tasks at scale. The pricing could make Muse Ember a practical choice for applications that require frequent AI operations.

Meta Muse Ember AI Model Is Built for AI Agents and Repetitive Tasks

Meta Muse Ember AI Model Is Built for AI Agents and Repetitive Tasks

Meta Muse Ember is designed to handle smaller tasks within larger AI agent workflows. It focuses on tasks that require frequent processing rather than complex reasoning.

The model can make tool calls, complete coding tasks, extract information, route requests, and process long documents. It can also analyze large collections of images, files, and records.

Muse Ember has a context window of up to 1,048,576 tokens and supports up to 524,288 output tokens. This allows developers to process large amounts of data in a single request.

The AI model accepts text, images, audio, video, and PDF files as input. However, it generates text-only responses.

Meta expects developers to use Muse Ember alongside larger models such as Muse Wonder and Muse Spark. These models can handle complex reasoning and planning, while Muse Ember manages smaller tasks across multiple operations.

For example, an AI agent could use a larger model to plan a customer support workflow. Muse Ember could then process individual customer queries, extract relevant details, and route requests to the right teams. This approach could help businesses run AI agents at scale while keeping operating costs low.

Meta Muse Ember Outperforms GPT-6 Luna and Claude Haiku 5.5 in Several AI Benchmarks

Meta claims its Muse Ember AI model can compete with rival models from Anthropic and OpenAI across several AI benchmarks. The company compared its performance with Anthropic’s Claude Haiku 5.5 and OpenAI’s GPT-6 Luna.

According to benchmark results released with the announcement, Muse Ember scored 36 on the Artificial Analysis Intelligence Index. Claude Haiku 5.5 also scored 36, while GPT-6 Luna recorded 28.

Muse Ember also delivered higher scores in several AI agent and coding tests. In AutomationBench-AA, which measures how well AI models complete business workflows, Muse Ember scored 37.8%. This was higher than GPT-6 Luna’s 20.2% and Claude Haiku 5.5’s 13.9%.

The model also scored 40.9% on Terminal-Bench 4.0. Claude Haiku 5.5 followed with 39.2%, while GPT-6 Luna scored 16.4%. However, Muse Ember did not lead every benchmark. Claude Haiku 5.5 performed better in FrontierCode 2.0-AA, scoring 34.9% compared with Muse Ember’s 27.2%.

The results show that Meta Muse Ember can compete with established AI models in several coding and AI agent evaluations. However, its performance varies depending on the benchmark and the type of task.

Meta Makes Muse Ember Available Through Its Model API and Muse Code

Meta has released Muse Ember through its Meta Model API and Muse Code. Developers can access the model using the model ID muse-ember-1.0.

The model also supports the Responses API and is available through Google Cloud and Oracle Cloud. This allows developers to integrate Muse Ember into their applications and AI agent workflows.

Meta has also added Muse Ember to its Muse Code subscription plans. Developers can use the model for coding tasks, automation, and other AI-based operations.

According to Meta, Muse Ember is the first publicly released model trained using Muse Wonder Full. Its release is part of the company’s wider effort to develop smaller and more efficient AI models.

Meta also assessed Muse Ember for potential cybersecurity, chemical, and biological risks. The company evaluated possible loss-of-control risks as well. It said the model remained below its high-risk thresholds across all assessed categories.

Meta Muse Ember Puts Pressure on OpenAI and Anthropic With Cheaper AI Pricing

Meta Muse Ember Puts Pressure on OpenAI and Anthropic With Cheaper AI Pricing

The launch of Muse Ember adds to the growing competition between Meta, OpenAI, and Anthropic in the AI market. Major AI companies are competing to build more capable models while reducing the cost of running them. 

This is important for businesses using AI agents, as a single workflow can require multiple model requests. These repeated requests can increase AI operating costs, especially when companies handle thousands of tasks daily.

Meta aims to address this challenge with Muse Ember’s low token pricing and focus on repetitive tasks. Its benchmark results also show that smaller AI models can deliver competitive performance in specific coding and automation tasks.

However, Muse Ember is not designed to replace larger AI models. Meta expects developers to use it alongside more advanced systems that handle complex reasoning and planning.

With Muse Ember, Meta is expanding its AI model lineup and competing on both performance and pricing. Its long-term success will depend on how well the model performs in real-world applications and whether businesses adopt it for large-scale AI agent workflows.

Posted in AI News | Tagged | Leave a comment

Anthropic Withdraws $69 Billion IPO Projections Sent to Banks After Acknowledging Errors

Anthropic has withdrawn its preliminary initial public offering (IPO) details after identifying errors in the proposed share pricing. The withdrawn figures suggested that Anthropic could raise up to $69.3 billion through its IPO. This could make it one of the largest stock market listings in history.

On November 9, 2026, Anthropic sent bankers a draft securities filing outlining plans to sell approximately 867 million shares at $70 to $80 per share. However, the company later informed bankers that the pricing information was incorrect. It also said that an updated IPO filing would be shared at a later date.

The withdrawal has left Anthropic’s IPO valuation, share price, and fundraising target uncertain. Investors are now waiting for Anthropic to release revised IPO details ahead of its planned stock market debut. The company faces growing scrutiny over AI valuations and rising infrastructure spending.

Anthropic Withdraws $69.3 Billion IPO Filing Over Pricing Errors

Anthropic has pulled back an initial IPO filing sent to investment bankers after identifying errors in its proposed share pricing details. According to Reuters, the document outlined plans to sell approximately 866.7 million Class A shares at an estimated price of $70 to $80 per share. At the upper end of this range, the Anthropic IPO could have raised approximately $69.3 billion.

The initial filing included shares from both Anthropic and its existing investors. The company planned to issue approximately 843.3 million new shares, while existing shareholders were expected to sell around 23.3 million shares.

However, Anthropic later informed the investment bankers that the share pricing details in the document were incorrect. The company did not specify which figures contained errors or explain how the incorrect information entered the filing. It also declined to comment publicly on the withdrawal.

Anthropic said it would send an updated IPO filing at a later date. Until then, the company’s proposed share price, total offering size, and final IPO details remain unconfirmed.

ALSO READ: Anthropic IPO Filing Reveals Huge AI Spending and Ambitious Growth Plans

Anthropic’s $26 Billion Private Placement Includes Major Tech Companies

Anthropic's $26 Billion Private Placement Includes Major Tech Companies

Anthropic’s withdrawn IPO filing also revealed plans for a separate $26 billion private placement involving several major technology companies. According to Reuters, Google’s parent company, Alphabet, Nvidia, Microsoft, AMD, Broadcom, Micron Technology, and TSMC had agreed to invest in the company.

Under the proposed arrangement, these companies would purchase Anthropic’s Class N common shares at the same price offered to investors in the public IPO. The private placement was expected to close immediately after the public share offering.

The proposed investment shows Anthropic’s financial ties with some of the biggest players in the technology industry. Several of these companies also provide the computing infrastructure, chips, and other resources that support Anthropic’s AI models and services. At the same time, some are developing their own AI products and competing in the growing artificial intelligence market.

However, the $26 billion private placement was outlined in the withdrawn IPO filing. Anthropic has yet to confirm whether the investment commitments, participating companies, and proposed terms will remain unchanged in its updated filing.

Anthropic Faces Valuation Pressure Ahead of Its IPO

Anthropic’s IPO plans come as investors become more cautious about the rising costs of artificial intelligence development and infrastructure. Heavy spending on data centres, AI chips, and computing resources has raised concerns about the long-term returns on AI investments.

According to Reuters, investors had previously discussed a potential Anthropic valuation of up to $2 trillion. However, expectations later declined, with $1.5 trillion considered the upper end of the expected valuation range.

A separate report by The Washington Post on November 10 said Anthropic was targeting an IPO valuation of more than $1 trillion. The company was also reportedly planning to sell $62.5 billion worth of shares, although the final terms could change following meetings with potential investors.

This means the previously reported $69.3 billion IPO figure is not Anthropic’s confirmed fundraising target. The company has yet to release updated offering details.

Rising interest rates and concerns about AI infrastructure spending have also affected technology stocks and bonds. Several companies in the AI infrastructure and data centre sectors have delayed or withdrawn their planned stock market listings.

Anthropic Reports $17 Billion Revenue Alongside $900 Billion Computing Commitments

Anthropic has reported strong revenue growth as demand for its Claude AI models continues to increase. According to its amended filing, the company generated nearly $17 billion in revenue during the three months ended September 30, 2026. However, its quarterly growth rate slowed sharply, raising questions about whether it can maintain the same pace of expansion.

Anthropic also has nearly $900 billion in future computing commitments with infrastructure partners, including Google and Amazon. These agreements cover the computing resources needed to develop, train, and operate its AI models.

The large commitments highlight the high cost of running advanced AI services. They also raise questions about Anthropic’s future spending as it expands its AI infrastructure and competes in the growing AI market.

ALSO READ: Anthropic’s $518 Billion AI Plan Locks In Years of Infrastructure Spending

Anthropic Prepares for IPO as Investors Await Updated Pricing Details

Anthropic Prepares for IPO as Investors Await Updated Pricing Details

Anthropic is preparing for its planned IPO, but investors are waiting for the company to release updated share pricing and offering details. According to Reuters, Anthropic was expected to begin its investor roadshow on November 10, with IPO pricing scheduled for the following week. 

During the investor roadshow, Anthropic will present its business, financial results, and growth plans to potential investors before setting the final IPO terms. The IPO will also test investor demand for major AI companies. 

Investors will closely examine Anthropic’s revenue growth, infrastructure costs, computing commitments, and future business prospects before deciding whether to participate in the share offering. Anthropic has not yet explained which figures in the withdrawn filing were incorrect. The final IPO share price, fundraising target, and offering details will depend on its updated disclosures.

Posted in AI News | Tagged | Leave a comment

Anthropic Launches Claude Fable 5.2 and Mythos 5.2 With Faster Performance

Anthropic has launched Claude Fable 5.2 and Claude Mythos 5.2, the latest versions of its advanced AI models. The new Claude models bring improvements in coding, reasoning, scientific research, and visual understanding. They also generate responses up to 30% faster than their previous versions.

Anthropic has also cut API prices by 20%, making its latest AI models cheaper for developers and businesses. However, access to Claude Mythos 5.2 remains restricted to verified organizations working in cybersecurity and life sciences.

The company announced both models on November 9, 2026. The launch comes as competition in the AI industry grows, with major technology companies investing heavily in advanced AI models, coding tools, and AI agents.

Claude Fable 5.2 Gets 30% Faster Performance and 20% Lower API Prices

Claude Fable 5.2 and Claude Mythos 5.2 use the same underlying AI model. However, Anthropic has introduced different safety restrictions and access limits for each version.

According to Anthropic, both models deliver responses up to 30% faster than their 5.1 versions. They also provide clearer responses and use fewer tokens for some complex tasks. Anthropic has also reduced the Claude Fable 5.2 API pricing by 20%, making the models less expensive for developers and businesses.

The updated API pricing is:

Input Tokens$8 per million tokens, down from $10
Output Tokens$40 per million tokens, down from $50

Anthropic has also added an optional Fast Mode for developers who need faster responses. This feature increases token generation speeds by up to 2.5 times. However, it comes at twice the standard API price.

The Fast Mode pricing is:

Input Tokens$16 per million tokens
Output Tokens$80 per million tokens

The lower API prices could make Claude 5.2 models more affordable for businesses and developers. Faster response times could also help companies handle complex coding, research, and other demanding AI tasks more efficiently.

ALSO READ: Anthropic Launches Claude Opus 5.5 With Fable-Level Performance at 60% Lower Cost

Claude Mythos 5.2 Access Limited to Verified Organizations

While Claude Fable 5.2 is available to a wider range of users, Anthropic has restricted access to Claude Mythos 5.2. The company is making the more powerful version available only to verified organizations.

Both models share the same underlying AI capabilities. However, Claude Mythos 5.2 has fewer restrictions on cybersecurity and biological research tasks. Organizations must pass Anthropic’s verification process to access the model. 

Cybersecurity organizations can apply through the Cyber Verification Program. Life sciences researchers can request access through the Life Sciences Verification Program. Anthropic has introduced these restrictions to reduce the risk of AI misuse. 

Advanced AI models could help users identify software vulnerabilities or conduct complex biological research. These capabilities could also create security risks if used for harmful purposes.

The company has also improved the safety filters in Claude Fable 5.2. According to Anthropic, its safety classifiers produce around 55% fewer unnecessary blocks than those in Fable 5.1. The improvements include:

  • Cybersecurity: Around 60% fewer unnecessary blocks.
  • Biology: Around 40% fewer unnecessary blocks.

These changes allow developers to complete legitimate technical and research tasks with fewer interruptions while keeping safeguards in place.

Claude Fable 5.2 Shows Major Improvements in Coding and Advanced Reasoning

Claude Fable 5.2 Shows Major Improvements in Coding and Advanced Reasoning

Anthropic says Claude Fable 5.2 and Claude Mythos 5.2 deliver major improvements in software development, scientific research, and complex problem-solving. During early testing, both AI models helped engineers complete complex software migrations. 

They also designed three working electronic devices from scratch and supported research into robotics. Anthropic also reported that the models solved two previously unsolved problems in black hole physics.

In another test, Claude Fable 5.2 helped researchers formalize a proof of Fermat’s Last Theorem in just 30 hours. The task involved using Lean, a mathematical programming language, to convert the proof into a format that computers could verify. 

The new Claude models also showed improvements in agentic coding and advanced reasoning. Anthropic reported the following benchmark results:

  • FrontierCode 2.0: Claude Fable 5.2 scored 49.1% at its highest reasoning level, compared with 45.5% for Claude Opus 5.5.
  • FrontierSWE v2: Fable 5.2 achieved 69.9%, while Opus 5.5 scored 62.3%.

These benchmark results show Claude Fable 5.2’s improved ability to handle complex coding and software engineering tasks. Its progress in mathematical reasoning and scientific research also points to broader improvements in advanced AI capabilities.

ALSO READ: Anthropic Unveils Claude Haiku 5.5 for High-Volume AI Tasks

Anthropic Expands Claude Fable 5.2 Access Across Major Cloud Platforms

Anthropic has made Claude Fable 5.2 available across its platforms, including the Claude API and major cloud providers such as Amazon Web Services (AWS), Google Cloud, and Microsoft Azure.

Developers can access the new AI model through the Claude Platform by using the model name claude-fable-5-2. Anthropic is also offering eligible paid subscribers additional Claude Fable 5.2 usage during the launch week. The extra usage will depend on their subscription plans.

However, Claude Mythos 5.2 remains available only to verified organizations. Anthropic has also introduced access through Claude Security for eligible Enterprise customers. The launch reflects Anthropic’s efforts to make its latest AI models faster and more affordable. 

The company is keeping stricter access controls on models with advanced cybersecurity and biological research capabilities. As competition in the AI market increases, Anthropic is taking a different approach to model availability. It is offering wider access to Claude Fable 5.2 while limiting access to Mythos 5.2 because of its more sensitive capabilities.

Posted in AI News | Tagged | Leave a comment