CMR is the leading provider of funding and management support for small to medium-sized businesses and entrepreneurs

Established 1984 CMR is the leading venture capital, management support and business services provider for small to medium-sized businesses - linking excellent management skills with the substantial financial resources of a global bank of private investors.
CMR has over 450 senior executives, operating in the UK, USA, Europe, Asia, Australasia and globally, providing both funding and specialist help for entrepreneurial businesses.

For Businesses CMR provides excellent resources:

CMR FundEX Business Exchange - gives all companies & entrepreneurs direct access to CMR's global investor base.
CMR Catalyst Group Programme - transform profitability through merging.
CMR Company Sales Division helps owners to exit at the best price.
CMR Corporate Recovery Division - experts in rescue and turnaround.
CMR Technology Licensing Division - commercialising innovation.
CMR Executive Professionals - management support and consultancy.
CMR Executives-on-Demand™ Fully experienced senior executives available quickly and cost effectively.

We always welcome contact with new business clients- please get in touch - we will do our best to match your needs and exceed your expectations.

For Investors

Preferential access to new opportunities for investment and/or acquisition
Pre-vets propositions and provides a personalised service to our investors
Syndication service enabling investors to link together as desired
Executive and management support for investments as needed
CMR's services to our investors are not only fast & efficient but also free

We always appreciate new members- you are welcome to join as an investor or as a CMR Executive.

When you join us as a Senior Executive:

CMR's strength is in the skills and experience of our executive members - all senior, director level people with years of successfully running and managing companies.
Because the demand for CMR's support and services is ever-increasing, especially as we enter recessionary times, we have a growing need for more high calibre executives to join us from every industry and discipline.
You will be using your considerable experience to help smaller businesses and entrepreneurs to grow profitably.
We offer full training and mentoring support to help maximise potential.

We are always keen to find more high calibre senior executives in all areas- skills and location. Make contact with us today and maximise your opportunities.

Venture Capital Funding

HEAD OFFICE

124 City Road

London EC1 2NX

Tel: +44 (0)207-636-1744

Fax:+44 (0)207-636-5639

Email: cmr@cmruk.com

Registered Office: 124 City Road, London EC1 2NX

Also Glasgow, Dublin, Switzerland, Europe, USA/Canada

Privacy Statement: CMR only retains personal details supplied directly by executives joining CMR themselves either as Full Executive Members or Interim Management Members or Investors. Those details are only used within CMR and not disclosed to any third parties without that person’s agreement. We will keep that data until requested by the person to be removed – at that point it will be deleted. Personal data is never sold or used for purposes outside of CMR’s normal operations. Any correspondence should be directed to the Managing Director, CMR, Kemp House, 152-160 City Road, London EC1V 2N

Senior Executives

CMR is a worldwide network of senior executives. Join us to expand your career and business horizons.

Business Entrepreneurs

CMR has a complete range of resources & services provided by experts to help all businesses to grow and prosper.

Investors & Venturers

CMR has a continuous stream of business and funding propositions, which are matched to investor preferences. Join us - it's FREE!

FundEX

FundEX is CMR's worldwide stock market for small to medium sized companies and entrepreneurs to raise new capital.

Interim & Permanent Management

Many of CMR's executives can be recruited on an interim, permanent or NED basis.

Main CMR Intranet members only

Regional Intranets

Mon, 20 Jul 2026 03:20:00 +0000

'Convergence': Secret Labs Already Deployed Lifelike Female Androids While Public Watches Optimus Toy Demos

'Convergence': Secret Labs Already Deployed Lifelike Female Androids While Public Watches Optimus Toy Demos

Authored by Madge Waggy, Read more.....

'Convergence': Secret Labs Already Deployed Lifelike Female Androids While Public Watches Optimus Toy Demos
Authored by Madge Waggy,

Technology has an unusual habit of revealing itself twice.

The first version appears quietly, almost unnoticed, inside research papers, patent filings, laboratory demonstrations and corporate presentations that attract little attention beyond specialists.

The second version arrives years later, polished into a consumer product that suddenly convinces everyone the breakthrough happened overnight. History repeatedly favors this illusion. By the time society begins discussing a technological revolution, thousands of engineers have already spent years solving problems the public never realized existed.

That gap between discovery and public awareness has become one of the defining characteristics of the twenty-first century, and nowhere is it more apparent than in the race to build machines that no longer resemble machines at all.

During the last three years alone, humanoid robotics has advanced at a pace few analysts considered realistic even a decade ago. Tesla continues developing Optimus as a general-purpose humanoid assistant, Figure AI has demonstrated robots capable of performing increasingly complex industrial tasks while collaborating with advanced language models, Boston Dynamics has introduced an entirely electric generation of Atlas after retiring one of the most recognizable research robots ever built, and companies such as Sanctuary AI, Agility Robotics and Apptronik are openly competing to place human-shaped machines inside factories, warehouses and commercial environments. Simultaneously, breakthroughs in computer vision, reinforcement learning, synthetic materials and multimodal artificial intelligence have dramatically reduced the distance between experimental prototypes and practical deployment. None of these developments belong to science fiction anymore. They are unfolding in plain sight, documented through academic publications, investor briefings and live demonstrations watched by millions around the world.

Yet an uncomfortable pattern has begun emerging alongside this extraordinary progress. Every major technological leap seems to generate a parallel conversation taking place far from universities and conference halls, unfolding instead across anonymous forums, encrypted chat groups and obscure corners of the internet where speculation often grows faster than evidence. Some theories disappear within hours because they collapse under the weight of obvious contradictions. Others survive for years despite the complete absence of proof, not because they successfully explain reality but because they ask questions that remain surprisingly difficult to dismiss. One recurring idea has become particularly persistent: if companies publicly demonstrate machines capable of walking, reasoning and manipulating objects with human-like dexterity today, what might exist inside research facilities whose work will not become public until five or even ten years from now?

No credible evidence has ever answered that question.

The absence of evidence, however, has never prevented the internet from trying.

Among the countless stories that circulate through speculative technology communities, one in particular continues resurfacing despite never being independently verified. It describes an invitation-only exhibition allegedly organized inside an anonymous research complex where visitors were shown humanoid entities unlike anything previously demonstrated in public. According to archived discussions that periodically reappear online, the event itself was almost deliberately ordinary. There were no theatrical presentations, no dramatic product launches and no executives standing beneath oversized screens announcing the future of artificial intelligence. The building resembled a research facility more than a convention center, its corridors illuminated with clinical precision and stripped of nearly every identifying feature that could reveal who financed or operated the installation. Whatever attracted attention that day was not the architecture. It was what stood behind the glass.

The descriptions vary in countless small details while remaining strangely consistent about the central image. Long rows of transparent containment chambers reportedly filled an immense exhibition hall, each enclosing a female humanoid whose appearance challenged the very definition of robotics. Visitors allegedly expected exposed actuators, composite frames or polished metallic joints similar to those displayed by contemporary engineering prototypes. Instead, they found figures possessing skin that reflected light with subtle biological imperfections, naturally distributed hair, nearly imperceptible facial asymmetry and eyes that appeared disturbingly capable of maintaining silent attention. Some accounts insisted they were sophisticated androids. Others argued they were advanced synthetic organisms. A few dismissed the entire story as an elaborate digital art project designed to exploit growing public fascination with artificial intelligence. None of these interpretations has ever been supported by verifiable evidence, yet none has managed to erase the story either.

Perhaps that persistence says less about the photographs than it does about the era in which they allegedly appeared.

Only a few years ago, generating a convincing human face required painstaking digital artistry. Today, artificial intelligence routinely produces images, voices and video sequences capable of deceiving experienced observers under ordinary viewing conditions. Simultaneously, roboticists are steadily solving challenges once considered decades away from practical implementation. Artificial muscle systems continue improving. Synthetic skin capable of sensing pressure and temperature is advancing through laboratories across Europe, Asia and North America. Neural networks increasingly interpret visual information with remarkable precision, while large language models have transformed natural conversation into something machines perform with astonishing fluency. Each breakthrough, taken individually, appears understandable. Together they create an unsettling possibility that public imagination often races ahead to explore long before reality catches up.

The modern conspiracy landscape thrives inside precisely that narrow space between demonstrated capability and undocumented possibility. It rarely invents technology from nothing. Instead, it observes genuine scientific progress, extends every trend several years into the future and asks whether society is already witnessing carefully selected fragments of a much larger picture. Throughout history, classified aviation projects, cryptographic systems and surveillance technologies have all existed years before their official acknowledgment. That historical precedent fuels endless speculation whenever another rapidly advancing field begins transforming the world. Humanoid robotics, perhaps more than any emerging technology today, naturally invites the same questions�not because hidden laboratories have been proven to exist, but because technological acceleration has repeatedly surprised even the experts attempting to measure it.

THE EXHIBIT THAT ARRIVED BEFORE THE FUTURE

Technology rarely introduces itself with a dramatic announcement. More often, it arrives disguised as another research paper, another patent application, another demonstration watched by a few thousand engineers before disappearing beneath the next day's headlines. Looking backward, every technological revolution appears inevitable. Looking forward, it almost always resembles coincidence. That contradiction has become increasingly difficult to ignore over the past several years as artificial intelligence and humanoid robotics have advanced at a pace that even optimistic forecasts struggled to anticipate. What seemed extraordinary in 2020 became commercially viable by 2024, and by 2025 discussions that once belonged exclusively to science fiction had quietly entered boardrooms, government agencies, and manufacturing plants around the world.

Unlike previous waves of automation, today's race is no longer centered around machines built exclusively for factories. The objective has shifted toward creating systems capable of operating naturally inside environments originally designed for people. That distinction changes everything. Companies such as Tesla continue refining�Optimus, Figure AI has demonstrated humanoid robots performing industrial tasks alongside advanced language models, Boston Dynamics has transitioned to an entirely electric version of�Atlas, while Sanctuary AI, Apptronik, and Agility Robotics are openly competing to deploy human-shaped machines into warehouses, logistics centers, hospitals, and commercial environments. None of these developments are hidden behind classified documents. They are documented through public demonstrations, engineering conferences, investor presentations, and peer-reviewed research. The world is not waiting for humanoid robots to arrive. They have already arrived; the only remaining question concerns�how quickly they become ordinary.

�
What makes this timeline remarkable is not any single breakthrough, but the compression between them. Technologies that previously evolved over decades are now progressing within months. Computing power continues expanding, neural networks become more efficient with every iteration, synthetic materials increasingly mimic biological tissue, and robotic dexterity improves through reinforcement learning systems trained inside simulated environments before ever touching the physical world. Several research groups are now experimenting with�electronic skin�capable of detecting pressure, temperature and texture, while others are developing artificial muscle fibers designed to reproduce the flexibility of biological movement. Independently, each achievement appears incremental. Viewed together, they begin resembling pieces of a much larger puzzle whose final image remains frustratingly incomplete.

Throughout history, transformative technologies have almost always existed in limited forms before becoming public knowledge. Stealth aircraft remained classified long before their official acknowledgment. Modern cryptographic systems evolved inside government programs years before entering consumer electronics. Satellite reconnaissance, advanced computing and even the internet itself all followed similar trajectories, transitioning gradually from restricted environments into everyday life. None of this proves that humanoid robotics follows the same path, yet it explains why discussions surrounding undisclosed research facilities continue appearing wherever technological acceleration outpaces public understanding.

Among the countless stories circulating across futurist forums, one recurring narrative refuses to disappear despite the complete absence of verifiable evidence. According to these accounts, anonymous images allegedly depicting a private exhibition began surfacing shortly after several major robotics announcements captured international attention. The photographs themselves were unremarkable at first glance. There were no dramatic explosions of light, no cinematic laboratories filled with exposed machinery and certainly no science-fiction aesthetic designed to impress an audience. Instead, they portrayed an environment whose greatest source of discomfort came from its overwhelming normality. Clean architectural lines, carefully controlled lighting, visitors quietly observing transparent containment chambers, and inside those chambers, female humanoids whose appearance blurred the distinction between biological life and engineered design so effectively that many viewers struggled to identify where one ended and the other supposedly began.

What fascinated online communities was not the imagery itself but the possibility it represented. If the photographs were fabricated, they reflected an extraordinary understanding of contemporary robotics and human anatomy. If they depicted an elaborate artistic installation, the creators had achieved precisely the emotional response they intended. Yet if�purely as a matter of speculation�they represented a glimpse into technology more advanced than anything publicly demonstrated, they suggested something profoundly unsettling about the pace at which artificial intelligence and embodied robotics might already be evolving beyond public awareness.

WHEN MACHINES STOP LOOKING LIKE MACHINES

Perhaps the most significant shift occurring today has little to do with processing power or mechanical engineering. It concerns�perception. Early industrial robots never attempted to resemble humans because efficiency mattered more than familiarity. Modern humanoid research pursues the opposite objective. Engineers increasingly recognize that machines designed to work alongside people benefit from recognizable gestures, natural movement and intuitive communication. Artificial intelligence no longer exists solely inside software; it is gradually acquiring a physical presence capable of navigating environments originally built for biological life. That transition introduces questions extending far beyond engineering specifications. It touches psychology, ethics, economics and identity itself.

Researchers studying the�uncanny valley�have documented for decades that human beings respond differently to machines once they become almost�but not entirely�indistinguishable from living people. Small imperfections that might otherwise pass unnoticed suddenly become psychologically significant. A smile held a fraction too long. Eyes that maintain uninterrupted focus without natural micro-adjustments. Facial muscles moving with perfect synchronization rather than subtle asymmetry. These details rarely appear frightening in isolation, yet together they produce an instinctive discomfort difficult to explain rationally. As synthetic materials improve and artificial intelligence becomes increasingly sophisticated, the boundary responsible for that sensation continues narrowing.

Whether humanity ultimately embraces lifelike humanoids or resists them remains impossible to predict. What can be stated with confidence is that the underlying technology continues advancing regardless of public opinion. Investment in embodied AI has accelerated dramatically, governments have begun drafting regulatory frameworks for autonomous systems, and major technology companies increasingly describe robotics as one of the next defining industries of the coming decade. Against that backdrop, it becomes easier to understand why fictional stories about hidden laboratories, invitation-only exhibitions and technologies existing several years ahead of public demonstrations continue capturing the imagination of millions.�

BEYOND THE GLASS

Whether the exhibition ever existed eventually became the least interesting question.

The discussions that followed took on a life of their own, gradually shifting away from anonymous photographs and toward something far more unsettling. Engineers began debating theoretical manufacturing limits. Neuroscientists questioned whether synthetic cognition would eventually require emotions rather than simply simulating them. Military analysts speculated about autonomous decision-making systems, while ethicists found themselves asking an entirely different question: if a machine could imitate every observable characteristic of human behavior, what objective measurement would still separate the creator from the creation?

Within the universe surrounding the alleged exhibition, researchers supposedly referred to the project using an unusual expression that later appeared across several archived discussion boards:�"Convergence."�The name carried no technical explanation, only a philosophical one. According to the mythology that slowly developed online, the objective had never been to construct machines capable of replacing human labor. That milestone had already become commercially achievable through ordinary automation. The real ambition was allegedly something considerably more difficult�to construct artificial beings capable of existing inside society without ever being recognized as artificial in the first place.

The concept itself had already escaped.

The documents discussed by online communities described laboratories unlike conventional research facilities. Engineers supposedly worked alongside psychologists, behavioral scientists, linguists, neurologists and artists, each responsible for solving a different aspect of the same impossible equation. Mechanical movement alone could never convince an observer they were looking at genuine life. Human beings unconsciously detect thousands of microscopic behavioral patterns every hour without realizing it. Eye contact lasts for predictable intervals. Facial muscles contract unevenly. Breathing changes with emotion. Tiny pauses interrupt ordinary conversation. Even silence possesses rhythm.�

An anonymous character allegedly summarized the problem in a single sentence that became strangely popular among the hidden communities based on these controversial topics:

"People don't recognize humanity because of perfection. They recognize it because perfection never survives long enough to become human."

THE FINAL PROTOTYPE WAS NEVER A MACHINE

As years passed, technological progress accelerated alongside real industry developments. Every genuine breakthrough in artificial intelligence strengthened public conviction regarding the rapid pace of innovation. Public demonstrations showed humanoid robots becoming smoother, quieter, and more capable each year. Language models learned to hold increasingly natural conversations. Synthetic voices lost their unmistakable mechanical cadence. Artificial vision systems improved beyond what many experts believed possible only a few years earlier. Observers interpreted every public announcement as confirmation that society was witnessing technologies whose foundational frameworks had already been quietly developed over the preceding years.

That progression reflected a psychological phenomenon repeatedly observed throughout technological history. Human imagination rarely invents futures from nothing. Instead, it extends visible trends until they become predictable realities. Steam engines became locomotives before becoming intercontinental railways. Primitive computers became smartphones through thousands of incremental improvements rather than one miraculous discovery. In much the same way, the Convergence Program mapped a future assembled not through impossible inventions, but through the gradual integration of disciplines already advancing today.

According to unpublished data, the exhibition itself was never intended to impress investors or government officials. It served another purpose entirely. Visitors unknowingly became participants in a behavioral experiment. Cameras hidden throughout the exhibition allegedly measured eye movement, hesitation, emotional response, interpersonal distance and unconscious reactions as guests walked past each observation chamber.

Every moment of curiosity became data. Every expression of uncertainty refined the next generation of synthetic behavior.

The humanoids behind the glass were not being evaluated.

The observers were.

That single twist transformed the entire legend from an ordinary tale into something psychologically far more disturbing. It suggested that humanity had misunderstood the experiment from the beginning.�

This ongoing focus on rapid technological evolution reflects a broader recognition that modern civilization is crossing a significant historical threshold with profound long-term implications. Artificial intelligence is actively transforming key sectors, including education, medicine, finance, manufacturing, and scientific research. Concurrently, humanoid robotics continues to advance at an accelerated pace, while brain-computer interfaces are successfully progressing through human clinical trials. Furthermore, synthetic materials are increasingly replicating the functional properties of biological tissue. While each of these advancements represents an isolated milestone, their combined integration shapes a highly complex and rapidly changing socio-technological landscape.

Tyler Durden Sun, 07/19/2026 - 23:20
Close

Mon, 20 Jul 2026 02:45:00 +0000

A Deep Dive Inside Kimi K3, And All Other Chinese AI Models: The Definitive China LLM Primer

A Deep Dive Inside Kimi K3, And All Other Chinese AI Models: The Definitive China LLM Primer

We were lucky enough to read the tea leaves ahead of the "frontier", so to speak, and conclude that various "open" models out of China woul Read more.....

A Deep Dive Inside Kimi K3, And All Other Chinese AI Models: The Definitive China LLM Primer

We were lucky enough to read the tea leaves ahead of the "frontier", so to speak, and conclude that various "open" models out of China would soon be the biggest talking point - not to mention the market's fulcrum catalyst. See for example:

More recently, last week we summarized virtually all of China's AI Models in "The Definitive LLM Primer" (July 11), which we republished below for our readers' convenience, yet in the fast-paced world of AI, even that article is now woefully out of data due to the recent arrival of Moonshot's open-sourced Kimi K3 (July 16), which has not only taken the AI world by storm, but promptly sparked a momentum meltdown amid growing fears that Chinese LLMs are catching up too fast to frontier US models (something we warned about one month ago here). The reason for the jarring market reaction is that Kimi K3 is viewed as being on par, if not better, than most leading US frontier models.

Big news: Kimi-K3 by @Kimi_Moonshot is now #1 in the Frontend Code Arena with 1679 pts, surpassing Claude Fable 5.

This is a 17-place jump from Kimi-k2.6 (#18 -> #1).

In Frontend, Kimi-K3 ranked #1 in 6 of 7 domains: Brand & Marketing, Reference-Based Design, Data & Analytics,� https://t.co/YDN3BufGkC pic.twitter.com/Oa6teaQnWp
� Arena.ai (@arena) July 16, 2026

And while there is rampant debate whether this performance is accurate or gamed to beat specific benchmarks, the bigger issue is that China is now clearly developing stunning(ly cheap) open-sourced models which are on par with much more expensive US models, which in itself renders the entire ROI calculus behind trillions in AI capex spending null and void, because if China can achieve 98% what the US has done with a fraction of the capex...

China is now leading the AI arms race after spending a tenth what the US did https://t.co/CkPbXzffwi pic.twitter.com/sJOwU0KZhp
� zerohedge (@zerohedge) July 17, 2026

... then all those trillions already allocated to AI capex are nothing more than sunk costs, as end markets will quickly pivot to what is cheapest, since in most cases it is also on par with what is best.�

So what happened for those who went on vacation early last week and are stunning how everything has changed?

Well, as Ronald Keung, the Goldman strategist who wrote the original Chinese LLM primer said late on Friday, we have gone "from cost efficiency (DeepSeek), rise in model intelligence (GLM) to new frontier/pricing power at Kimi K3."

Below we excerpt from his latest note (available to pro subs):

On July 17, the Kimi K3 open-weight model was released with 2.8 trillion parameters, and has set a new frontier in coding and certain agentic capabilities globally, as per Arena.ai coding rank and Artificial Analysis Intelligence score.

Accordingly, Goldman notes K3�s pricing was set at a new high amongst Chinese models, US$2.3 per 1M tokens (blended) vs. Qwen3.7 Max of US$1.4/Zhipu�s GLM5.2 of US$0.9/MiniMax�s M3 of US$0.22/DeepSeek�s V4 Pro of US$0.18, being still below global SOTA models.

As highlighted in our recent China LLM primer, China�s AI open-source/open-weight models are reaching a critical point of intelligence performance for global proliferation. Amongst signposts into 2H 2026, we have anticipated that competition could intensify in the high-end coding segment via coding data flywheel and scaling (to larger models, up to 2-5 trillions parameter sizes). The share price reaction of Knowledge Atlas/Zhipu (where Goldman recently initiated at Neutral, -28% on July 17) and MiniMax (Buy-rated, -16% on July 17) have been due to concerns around Chinese AI model competition and who the potential long-term winners will be, given the competitive landscape/sustainability of model company leadership remains highly dynamic.

Here, Goldman repeats that Independent AI model companies stand out in its Competitive Positioning framework (see full report for more). The bank also notes the positive read from Chinese President Xi�s comments at the opening ceremony of World Artificial Intelligence Conference (WAIC) in Shanghai held last Friday, drawing parallels of AI�s societal impact to the invention of electricity and steam engines, and in offering China�s technology infrastructure to developing nations with its continued open approach, yet cautioned on the need for human oversight/controls and risk mitigation.

What to watch out for from here?�

Harness/agentic applications: We expect China AI model companies to increasingly focus on positioning their harness/agentic applications as key entry points, especially in coding (e.g. Zhipu�s ZCode, Tencent�s Workbuddy, and Alibaba�s Qoder which are aggregate platforms that support the full suite of AI models) as model companies attempt to close the loop in capturing more real-life coding and agentic data for scaling. Co-work and industry expert agentic products could be the next priorities.
Multiple large parameter high-end coding model launches and potentially�stricter access to the most advanced Chinese AI models outside of China: After Kimi K3 launch, we anticipate for further new Chinese AI model launches over 2H26 (Zhipu GLM, Alibaba Qwen, MiniMax M3 Pro and more) with significantly larger total parameter sizes of 2-5 trillion across Chinese AI model players. Coding/programming segment competition will remain intense. The addition of multi-modal/visual understanding will also be the next upgrades amongst Chinese AI foundation models.
Continued suppressed API pricing for the lower end agentic-focused segment:�Goldman expects API pricing and therefore gross margins to remain under pressure at the lower-end pricing segment (around US$0.1-0.2 per 1M tokens) into the second half, as China�s AI model players have significant cash buffers post-fund raising to subsidize competitive pricing. As a result, we see financial strength as one of the three most important metrics (alongside pricing power and cost efficiencies) in assessing a Chinese AI model company.
Multi-modal/video-generation models to see further ARR ramp-up from global adoption: Goldman expects continued healthy industry pricing and gross margins within video generation (unlike foundation text) where key players ByteDance�s SeeDance (at reportedly 70% gross margins, latest US$2bn ARR run-rate), Kuaishou�s (Buy) Kling and MiniMax�s (Buy) Hailuo/upcoming H3 models to enjoy healthy growth over 2H 2026 amid new functionality breakthroughs (combination of video-generation with LLM) and tight computing resources where demand significantly outpaces capacity.
Expect increasing domestic ASICs supply, any stricter access to China�s�future most frontier models for overseas markets, Western markets� policies on China models, and access to high-end computing in model training as key swing factors/risks to our China AI model token/revenue growth trajectory.

Goldman's key ideas/stock picks

Within Cloud & Data Centers (the bank's top preferred sub-sector within China Internet),�GS continues to highlight key ideas (Alibaba, GDS, VNET, and Kingsoft Cloud) on the back of higher AI hyperscaler capex spending into 2H 2026.

Within AI models, Keung highlights MiniMax on upside skewed risk-reward. Key swing�factors for MiniMax to improve its overall competitive positioning will hinge on its pricing power following its recent completed fund-raising that has strengthened financial strength. Expect its H3 video generation model performance (imminent launch in a more favorable industry landscape vs. text models) and time-to-market�for its next M3 updates (focusing on further coding intelligence level uplift from post-training/reinforcement learning, we estimate M3 update over July-Aug, and a larger parameter size M3 Pro model later in 2H 2026) will be the next key drivers.

Yet what may be the clearest signal yet just how competitive with the US Chinese LLMs have become comes from none other than OpenAI's "head of strategic futures" (presumable that title refers to whoever can demand protectionism the loudest) who essentially is begging for protectionism against open Chinese LLMs...

Some observations on Kimi:

1. It's a very good model! I don't think its performance can be explained away by distillation or anything like that. In agentic coding sessions, it seems pretty much on par with the best public models of Q1 2026. In my fairly limited use, it also�
� Dean W. Ball (@deanwball) July 17, 2026

... and the scathing retorts from Trump's (former?) AI Tsar David Sacks, who responds that "K3 just fixed 15 critical security bugs that Codex and Fable refused because of �cyber guardrails.� There�s no reason to limit American models on tasks that Chinese models handle without issue. We�re only making ourselves less competitive." And then again, here to point out that Chinese LLMs are already more useful to some programmers than the best of the (very expensive) best Anthropic and OpenAI have the offer, to wit: "Here�s another example: Hugging Face tried using American frontier models to analyze an AI-powered cyber attack. But the guardrails blocked requests containing real exploit payloads so they switched to GLM 5.2 running locally. The guardrails actually impaired defensive security."

More in the full Goldman China LLM update focusing on Kimi K3, available to pro subs.�

Meanwhile, for those who missed our original report on Chinese LLMs�from July 11, we republish it below in its entirety.�

* * *�

Three weeks ago, we attempted a lengthy answer of the "trillion dollar question" namely are Chinese AI models a better value than US models�and, using extensive research from UBS, concluded that at almost 95% of the capability (and rising) and just 10% of the cost, the answer was a resounding yes.

Fast forward to today when Goldman analyst Ronald Keung also addressed the $64 trillion elephant in the room, and published�a 50-page China AI models LLM primer (available to pro subs), in which he agrees with our conclusion, namely that "China's AI open-source/open-weight models are reaching a critical point of intelligence performance vs. global proprietary models, with a significant ramp up in domestic enterprise & global SME adoption that will enable a positive data flywheel of further model improvement."�

In the report, Goldman evaluates:

How these models achieve such performance at low costs/tight computing resources;
Why they pursue an open-source/open-weight approach and how they monetize;
What the key addressable markets are, as global enterprises shift from 'token-maxxing' to ROI-first, where Goldman highlights two favored ARR quadrants for Chinese models; and
Who the potential long-term winners will be under the bank's Competitive Positioning framework.

In keeping with the recent newsflow, Ronald notes that ongoing politicization of AI, namely any stricter access to China's future 'most frontier' models for overseas markets, Western markets' policies on China models, and access to high-end computing in model training, are three key swing factors/risks to the bank's China AI model token/revenue growth trajectory.�

Additionally, the Goldman strategist introduces his Chinese AI model Competitive Positioning framework based on pricing power, cost advantage and financial strength, overlaying with token scale and market share progressions, and identify Knowledge Atlas (Zhipu, initiation) and DeepSeek (private) as the strongest positioned in foundation models, and��Bytedance (private) in multi-modal.

Before we get into the weeds, excerpted from the gull Goldman report, lets start with a visual summary of China's AI Model and Hyperscaler Ecosystem...

... and�a Breakdown of China's open-source models unit economics today and path to profitability

Which brings us to the core overarching theme of Chinese AI development, from cheap to smart, or as Goldman's Ronald Keung�puts it:�

From DeepSeek�s moment last year (on cost efficiency) to Zhipu�s GLM moment this year (on model intelligence).

China�s AI open-source/open-weight models are reaching a critical point of intelligence performance vs. global proprietary models, with a significant ramp up in domestic enterprise adoption and a proliferation of global consumer and SME demand. In this report, Goldman evaluates:

How these models achieve such performance at low costs/tight computing resources;
Why they pursue an open source/open weight approach and how they monetize;
What the key addressable markets are, where the bank highlights two favored ARR quadrants and risk factors; and
Who the potential long-term winners will be under Goldman's Competitive Positioning framework.

Chinese models are reaching a critical �good enough� stage for agentic tasks/specific coding scenarios, and rising fragmentation in China�s AI model landscape (but the strong will get stronger). While pricing power remains strong for models with frontier performance/multi-modal, the lower-end segment is in a price war; nevertheless, Agentic AI is driving explosive demand for these value-for-money models at the lower-end. Access to computing will be a swing factor, where US/China regulations, balance sheet and inference efficiency are key. Accordingly, the bank introduces its Chinese AI model Competitive Positioning framework based on pricing power, cost advantage and financial strength, overlaying with token scale and market share progressions, and identify several companies its views as the strongest positioned in both foundation models and in multi-modal.

How do Chinese models achieve competitive performance at low costs/tight computing resources? By leveraging smaller-sized parameter models for equivalent benchmark performance, and Mixture-of-Expert and new architectural innovations. Chinese AI models are bifurcating into a two-tiered market where performance and time to market are key to pricing power, thereby leading to two �ARR maximizing� quadrants based on token adoption and pricing. A positive flywheel is taking effect for top Chinese AI models driven by increasing actual real world coding adoption, and reducing their reliance on model distillation practices.

Why are Chinese models pursuing an open source/open weight approach, and ways to monetize? Open source allows for greater flexibility in model training/deployment, and allows for the widest adoption and an open community. Open-source models� disclosed ARRs are likely understating total deployment and revenue potentials, and Goldman expects more shifts to open weight (with Community License, i.e., commercial terms if for commercial use) among Chinese AI models down the road.

What are the key addressable markets, domestically and internationally, and key risks? The bank highlights two favored ARR quadrants and estimate China/China AI models market to see token growth of 25X by 2030E; the coding landscape�to consolidate while the agentic/low-end segment could remain fragmented. International (going global) should be a key upside, with the potential for higher pricing and global proliferation, especially in non-US markets as global enterprises increasingly pivot from a token-maxxing to a ROI-first model that prioritizes clear task boundaries, number of agents per day, back-end process automation and actual output over pure computational token volume.

Key risk factors are market access/anti-distillation and regulations, including any stricter access to China�s future most frontier models for overseas markets, access to highest-end leased computing equipment (used for training), market access to western markets, further restriction lists/entity list designations (but this could be positive for China�s path to further AI self-sufficiency across software/CPU/ASICs), and competition from SLM/threat from AI architectures.

Who are best positioned to be the long term winners within China�s AI model companies? Goldman expects players with the largest ARR scale with a gross margin advantage + financial strength to be the long-term winners. The bank highlights independent AI model companies mostly stand out in its Competitive Positioning framework in pricing power + cost advantage (in aggregate represent over US$200bn in implied valuations, based on latest market cap/funding rounds), where Zhipu and DeepSeek are the most strongly positioned in text based foundation models, while ByteDance leads in multi-modal capabilities.

Signposts for 2H 2026

Harness/agentic applications: China AI model companies will increasingly focus on positioning their harness/agentic applications as key entry points, especially in coding (e.g. Zhipu�s ZCode, Tencent�s Workbuddy, and Alibaba�s Qoder which are aggregate platforms that support the full suite of AI models) as model companies attempt to close the loop in capturing more real-life coding and agentic data for scaling. Co-work and industry expert agentic products could be the next priorities. Enterprises will increasingly focus on overall cost per task instead of headline pricing per token, with an openness to using different models (multiple-models approach).

Multiple large parameter high-end coding model launches and potentially stricter access to the most advanced�Chinese AI models outside of China: Multiple new Chinese AI model launches are expected over 2H26 with significantly larger total parameter sizes of 2-5 trillion across Chinese AI model players. Coding/programming segment competition will intensify as Chinese models try to challenge Zhipu GLM�s leadership via training on high-quality real-life coding data (where available) and scaling to larger parameter model sizes. There may be a potential shift from an open-source to an open-weight approach for best-performing models (i.e. from free-for-all use cases to requiring revenue sharing/a take rate for commercial use). The addition of multi-modal/visual understanding will be the next upgrades amongst Chinese AI foundation models (e.g. for GLM, DeepSeek), while MiniMax M3 already excels in these given the multi-modal focus of MiniMax from the start. That said, press reports on potential future restrictions on overseas access to China�s most advanced AI models (both closed and open source if at the frontier level) as cutting-edge artificial intelligence is increasingly being seen as a critical national asset.

Continued suppressed API pricing for the lower end agentic-focused segment: While DeepSeek recently announced an increase in peak-hour pricing from mid-July, API pricing and therefore gross margins should remain under pressure at the lower-end pricing segment (around US$0.1-0.2 per 1M tokens) into the second half, as China�s AI model players have significant cash buffers post-fund raising to subsidize competitive pricing at zero/negative gross margins in the near term. As a result, financial strength will be one of the three most important metrics (alongside pricing power and cost efficiencies) in assessing a Chinese AI model company, where cash on hand, net cash as % of assets and valuation multiples will be the critical financial strength metrics for long-term success. This said, Goldman is Buy-rated on MiniMax as the company stands out on cost efficiency/cost advantage metrics under a Competitive Positioning framework. With its M3 model well positioned in the favored ARR maximizing quadrant (attractive pricing + high token volumes), alongside its discounted valuation at 13X P/2026E year-end ARR (vs. China/global peers which command multiples several times higher at similar ARR stage), risk-reward is skewed to the upside (Goldmanb reiterates its Buy rating of the stock). Key swing factors for MiniMax to improve its overall competitive positioning will hinge on its pricing power and financial strength. Its H3 video generation model performance (imminent launch in a more favorable industry landscape vs. text models) and time-to-market for its next M3 updates (focusing on further coding intelligence level uplift from post-training/reinforcement learning, estimated over July-Aug, and a larger parameter size M3 model later in 2H 2026) will be the next key drivers.

Multi-modal/video-generation models to see further ARR ramp-up from global adoption: Goldman expects continued healthy industry pricing and gross margins within video generation (unlike foundation text) where key players ByteDance�s SeeDance, Kuaishou�s (Buy) Kling and MiniMax�s Hailuo/upcoming H3 models to enjoy healthy growth over 2H 2026 amid new functionality breakthroughs (combination of video-generation with LLM) and tight computing resources where demand significantly outpaces capacity. According to China news reports like LatePost and 36Kr, ByteDance�s Seedance gross margins have been at a healthy 70% at its latest US$2bn+ ARR run-rate.

The bank's strategists also expect increasing domestic ASICs supply, tighter access to overseas computing resources and potential market�access limitations (could mirror TikTok�s trajectory where rapid expansion in western markets was followed by more regulations/focus on ensuring data security where computing will have to be conducted within local jurisdictions).

Goldman's assessment of AI model companies: ARR scale x gross margin advantage + financial strength

Largest ARR scale (Token scale x pricing power)
Gross margin advantage (Training & Inference efficiency, Technology)
Financial strength (Balance sheet, Access to computing)

Accordingly, the bank lays out its Competitive Positioning framework for AI model players based on quantifiable metrics, 1) Pricing Power (amongst which, based on Time-to-market of model launch, Arena score based on actual usage cases and pricing�level), 2) Cost advantage (based on token volume scale, throughput/cache hit rate, parameter sizes/activation ratio and our estimate of inference gross margins), and 3) Financial strength (based on cash on-hand, net cash as % of assets, and valuation multiples).

Next, an overview of competitive analysis for key LLM labs� positioning�

The bank identifies two favorable ARR quadrants (in maximizing ARR) which is a combination of maximizing token volumes and/or pricing level

Comparing China�s key LLM players

Decoding China�s AI model tokens & our forecasts on token/revenue share

How do Chinese models achieve competitive performance at low costs/tight computing resources

Smaller-sized parameters models for equivalent benchmark performances, Mixture-of-Expert and new architectural innovations

As readers of our Chinese LLM primer may recall, compared with a year ago, Chinese models� coding and agentic capabilities have reached a critical level in being able to complete more coding and autonomous agentic tasks with higher success rates as context windows have been expanded to 1 million tokens. The smaller parameter model sizes of Chinese models (spanning from 200bn to 1.6T parameters, at 2-10% of leading SOTA models, due to constrained access to high-end computing), and highly efficient architectures (MoE, Sparse Attention, OCR etc., at 3-5% activated parameters only vs. total parameter sizes) all contribute to the much lower training and inference costs for Chinese models vs. leading US models. Goldman attributes the recent step improvement of Chinese models in coding according to Arena.ai to data curation, distillation techniques and reinforcement learning post training, despite their relatively small parameter model sizes (1.6T for DeepSeek V4 Pro, 0.7T for Zhipu�s GLM5.2 and 0.4T for MiniMax�s M3). On June 27, DeepSeek introduced DSpark, a speculative decoding framework that makes existing DeepSeek-V4 models serve faster. DSpark has already been deployed in DeepSeek-V4 Flash / Pro online serving, improving per-user DeepSeek-V4 generation 60-85% faster on V4-Flash and 57-78% faster on V4 Pro without changing the model�s weights or output quality.

Chinese AI models bifurcating into two-tiered market (where performance and time to market are key to pricing power), with two �ARR maximising� quadrants based on token adoption and pricing

Goldman is seeing pricing power for the highest performing Chinese AI models, e.g. Zhipu�s GLM5.2 model and Alibaba�s Qwen3.7 Max models at around US$1 per blended 1M tokens, at 5X that of low-end Chinese AI models. The reported tighter US processes in allowing most SOTA model access has also opened new arenas for China�s top performing coding models for enterprises and SMEs. Smaller parameter and activated ratios allow China�s top performing models to be priced at US$1, 10-25% vs. US SOTA models at US$4-8 per blended 1M tokens, while generating double digit 10-20% gross margins (GSe) that are lower than global SOTA models due to relatively lower pricing power.�In the lower-end segment, agentic focused models are priced at US$0.06-0.2 per blended 1M tokens, which are enabling these models to tap into new global TAMs for price sensitive SMEs and one-man companies. MiniMax generates 60-70% revenues from overseas.�

Note that DeepSeek announced that its V4 official version is set for launch in mid-July, alongside the introduction of peak/off-peak API pricing to better allocate resources and improve service stability. V4 Pro/Flash non-peak pricing remains unchanged, while peak hour (9am-12pm/2pm-6pm China time) will be charged at 2X non-peak rates, implying blended�pricing of US$0.35/US$0.12 per 1M tokens due to strong Chinese AI model demand that is increasingly causing significant compute tightness in work/productivity scenarios.

Positive flywheel is taking effect for top Chinese AI models from increasing actual real world coding adoption, with less reliance on model distillation ahead

Compared with learning and distillation tactics from global SOTA models in the past, top Chinese models like GLM5 are reaching a critical stage of adoption by China�s major enterprises and global users. As per LatePost, AI-generated code has increased to as high as 90% at some China mega-cap companies, up from 20-30% in 2H25, which will enable a positive data flywheel effect of further improvements via reinforcement learning with actual user data (both successful and unsuccessful coding cases) and post training. These have underpinned the step improvements in GLM5.2 from GLM5.1 in just over the course of a few months in 2026, and expect to see further step improvements to Chinese AI models over the next 6-12 months.

In coding/agentic tasks, Chinese players are reaching global top-tier positions�

Compared with global SOTA leading models of several tens of trillions, China open source models are mostly around or below 1 trillion parameter in size, and adopt a MoE structure, with low activated to total parameter ratios for higher inference efficiency.

Blended LLM token price (SDLLMTK) has been declining since early June, potentially driven by rising adoption of cost effective China models

Historical and projected capex for major US & China cloud service providers

Capex to operating cash flow ratio is still healthy for China hyperscalers

Case study on Meituan�s LongCat 2.0: A milestone for China�s domestic AI infrastructure

Released on June 30, 2026, Meituan�s LongCat 2.0 marks a major milestone as China�s first official 1.6 trillion-parameter open-source Mixture-of-Experts (MoE) model trained and deployed entirely on a 50,000-card domestic compute cluster. Purpose-built for agentic coding and complex software engineering workflows, the model features a native 1-million-token context window enabled by LongCat Sparse Attention (LSA) and dynamically activates an average of 48 billion parameters per token to optimize inference costs. Implications for a more self-sustainable China AI model outlook that is less reliant on foreign high-end chips for model training: The successful end-to-end pre-training and inference of a trillion-parameter class model on Chinese silicon (reportedly utilizing Huawei Atlas-950 SuperPods) fundamentally drives a more sustainable China�s AI model development outlook, in our view. While previous Chinese flagship models, such as DeepSeek V4-pro, mentioned domestic chips for inference, LongCat 2.0�s ability to overcome critical memory bottlenecks�and distributed stability challenges during the compute-heavy pre-training phase proves the viability of a wider and more localized hardware stack for AI model training in the future.

Why are Chinese models pursuing an open source/open weight approach, and ways to monetize?

Open source allows for higher flexibility in model training/deployment, and allows for the widest adoption and an open community�

Alibaba�s Qwen model family has long pursued an open source approach (before adopting a closed source for its largest and highest performing Qwen-Max models for better monetization), while other key Chinese AI model players have mostly pursued an open source/open weight approach including DeepSeek, Zhipu�s GLM and MiniMax M3 series models with the exception of ByteDance�s full closed proprietary approach for its Seed model. The open source approach allows for the highest flexibility in terms of the locations of where models are trained (and thus where the models can be deployed both inside and outside of Mainland China after training). An open source approach also allows for the highest adoption amongst the AI community with full transparency of the model parameters/architecture for trust, and an open community in driving more user feedback and thus model iterations/improvements. The open source approach vs. world�s leading closed/proprietary approach also provides an alternative choice for worldwide AI users when the best performing closed models have stricter user access and higher costs from their premium pricing, especially at a time when �token-maxxing� has become a key cost consideration for many corporates.

Open source models� disclosed ARRs are likely understating total deployment and revenue potentials

While open source model companies offer their own coding plans and a chargeable open platform API channel (where the model companies conduct their own model inference), the majority of open source models also allow individuals/third-party hyperscalers/neocloud providers to deploy the models without a charge even for commercial use (e.g. Alibaba Cloud�s Bailian MaaS platform can house GLM5.2 open source model without needing to pay a fee/take rate to Zhipu). As a result, while Zhipu�s last stated ARR target for year-end 2026 is at US$1bn, the actual deployment of GLM5.2 model worldwide is and will be multiple-fold higher vs. Zhipu�s own API channel token volumes and revenue. There is also increasing post training of Chinese AI models that are re-branded by global companies (e.g. Composer 2 etc.) where the Chinese AI models do not necessarily receive any revenues given the open source spirit.

Expect more shifts to open weight (with Community License) approach among Chinese AI models down the road

While Zhipu�s open source GLM model�s MIT license allows for free for all use cases (regardless of revenue), MiniMax M series models have pursed a restricted license (where the industry terms it as an open weight with Community License model), which requires MiniMax�s agreement and commercial terms (e.g. revenue sharing/a take rate) on commercial use. This will be the likely next path for other open source models in the Chinese AI model industry, in order for eventual gross profits of inference tokens to cover training costs and for each AI model company to achieve a sustainable path to returns.

What are the key addressable markets, domestically and internationally, and key risks?

According to Goldman estimates, China AI models� aggregate API+subscription revenue to increase from Rmb35bn in 2026E to Rmb879bn by 2030E from rising model intelligence, in particular with recent models like GLM5.2 reaching a critical point for global adoption and attractive pricing. The bank's revenue pool estimates for China AI models imply total Chinese AI model daily token consumption of 350T in 2026E to increase to 4,600T by 2030E.

Goldman estimates domestic market to see token growth of 25X by 2030E; coding landscape to consolidate while agentic/low-end segment could remain fragmented.

At 140tn daily token volumes for the country in March 2026 and several hundred trillions by June 2026 as per the National Bureau of Statistics, open source/open weight models have roughly a 30% token share (vs. 70% token market share by ByteDance alone, which is closed source and mainly driven by its Doubao app enterprise and individual user base, as the #1 used AI chatbot in China). Similar to the US, the coding segment (at a premium pricing level) will continue to be dominated by SOTA best performing models. Meanwhile, the lower-end segment focused on agentic AI will remain fragmented with multiple players due to the financial strength of AI model companies/mega-caps which would sustain the lower-end segment price war for longer.

International (going global) to be the key upside; with potential for higher pricing and global proliferation, especially in non-US markets

Goldman's US research team estimates agentic AI will drive 24X growth in token consumption by 2030 (from 2026) to 120 quadrillion tokens per month (or 4 quadrillion tokens per day, from their estimate of 170 trillion daily tokens today), with the biggest driver at 55X from enterprise agents and 12X from consumer agents. The global (ex. China) landscape has seen significant token share gains from Chinese AI models, as rising model intelligence and attractive token costs have driven higher adoption across 24/7 Hermes/Claw/Productivity agents, and shifting global SME mindset on using Chinese models in managing token costs given Chinese models have reached a �good enough� stage in terms of intelligence/performance.

Pivoting from �token-maxxing� to ROI-focused metrics, e.g. Daily Active Agents/Agentic Work Units

The AI token proliferation is undergoing a paradigm shift from �token-maxxing� (an initial focus from late 2025 to early 2026 where enterprises equated high AI token consumption directly with organization productivity) towards an �ROI-first� model that prioritizes clear task boundaries and output over raw computational volume.

�Token-maxxing� has been attributed to corporate inefficiencies and cost overruns: Data from a Jellyfish AI Engineering trends study indicated heavy AI users at enterprises consumed 10X more tokens but only had a 2X increase in output. Meanwhile, multiple US Internet companies have commented back in April 2026 that either their engineering teams utilized a full year�s AI budget in just four months using agentic products (promoting stricter monthly caps per tool) or changed their internal token utilization leaderboards that previously wrongly incentivized staff to launch inefficient/low value autonomous agent tasks.
Enterprise framework is pivoting towards an ROI-first model that prioritizes clear task boundaries, number of agents per day, backend process automation and actual output over pure computational token volume. Besides a marked increase in adoption of Chinese AI models, global enterprises have been downgrading default models to cheaper flash models for more typical tasks (while reserving SOTA models for only the top most value creating tasks like coding/programming). There is a transition from just token tracking to metrics like Daily Active Agents (DAA) and Agentic Work Units (AWU), and the overall cost per task will become more relevant than price per token metrics.

Open source/open weight approach allows for the option for U.S. hyperscalers to host Chinese models, operated within U.S. cloud ecosystem

Alphabet and Amazon�s respective cloud services Gemini Enterprise Agent Platform (via. its Model Garden) and AWS Bedrock already offer a broad selection of Chinese AI models including DeepSeek, MiniMax, Moonshot, GLM and Qwen which are fully managed by the US hyperscalers. Besides the model layer, into applications, worthy of note is Microsoft (covered by Gabriela Borges) CEO�s recent remarks at a Wall Street Journal interview (link) where he noted Microsoft is considering hosting versions of DeepSeek on Copilot as an optional, cost-effective model which could give its customers access to cheaper choices alongside US proprietary models. Microsoft indicated that if it hosts DeepSeek, the model would operate within its cloud ecosystem, ensuring customer data stays inside Azure.

Noting key risks around any potential tighter geopolitical policies given Chinese AI models inroads into western markets.

Key risks to the �going global� opportunity will hinge on end market access (especially in western countries, and with a focus on where computing is done/data is stored), access to high-end computing for AI model training (that could impact iteration pace and cost structure of Chinese models), and restrictions on Chinese AI model companies could impact supplier relationships/access to US companies and/or access to US capital.�

Who are best positioned to be the long term winners? Introducing our Competitive Positioning framework

Players with the largest ARR scale with a gross margin advantage + financial strength are expected to be the long-term winners.

ARR scale x gross margin advantage + financial strength

Largest ARR scale (Token scale x pricing power)
Gross margin advantage (Training & Inference efficiency, Technology)
Financial strength (Balance sheet, Access to computing)

Accordingly, Goldman introduces a Chinese AI model Competitive Positioning framework based on pricing power, cost advantage and finance strength, overlaying with token scale and market share progressions, and identify Knowledge Atlas (Zhipu initiation) and DeepSeek (private) as the most strongly positioned in foundation models, and Bytedance (private) in multi-modal capabilities

Foundation models

Goldman assesses each key player�s competitive positioning from its flagship foundation model, scored across 3 aspects: pricing power, cost advantage and financial strength (of the company), with each aspect further built upon granular quantitative metrics.

Pricing Power

Time to market:�Goldman assesses this in terms of how quickly and effectively a player delivers frontier competitive models. From comparing the launch date of the company�s flagship models with its prior generation & other models at similar performance level, Goldman assesses whether the release delivers a meaningful capability step-up over its past generation, and whether it narrows the gap to the current global SOTA models.

Arena score (overall text): Model intelligence is viewed as the primary determinant of pricing power, since models capable of handling higher-value tasks can deliver clearer ROI and therefore sustain their pricing premium. Here, the LMArena�s score is used instead of any static benchmark because it reflects a large scale of blind user reviews, making it a more objective read on the real-world capability.

Pricing (blended, US$ per 1M tokens): The bank�refer to the realized headline price (across input/output/cached input) for flagship models, where sustained higher pricing with iteration signals stronger pricing power, whereas pricing cuts may indicate a more volume-prioritized strategy.

Cost advantage: the structural cost-to-serve (i.e. inference costs) that sets the floor for price and margin

The below metrics are referred to as proxies for inference efficiency:

Throughput (tokens/second): Measured by the number of tokens a single GPU generates per second. The more tokens a GPU outputs per second, the more fixed hourly compute cost is spread out. Therefore, throughput is a strong indicator of inference efficiency, which depends on model architecture (sparsity & attention design) and serving efficiency (batching & utilization).
Cache hit rate (%): The share of input tokens served from cache rather than recomputed. In conversations and agentic workflows, much of the input repeats across calls, so models can read those tokens from cache memory instead of�recomputing from new inputs. A higher rate cuts compute, and since cached tokens are near-free to serve despite being billed at a discount (10X-100X cheaper than input cost), it is also margin accretive.
Parameter size/activation ratio: The share of total parameters activated per token (available for open-source MoE models), where fewer active parameters mean fewer FLOPs per token and therefore lower inference cost, at any given level of performance.�
Inference GPM (where disclosed, or GS estimates based on total parameter size/activated parameters): The realized gross margin on model API, where ~90% of COGS is inference costs, as a direct indication on cost efficiency

Financial strength: capacity to keep funding frontier R&D and training before a profitability turnaround

Cash on hand and net cash/debt as % of assets: Goldman uses total cash on hand as a measure of balance sheet strength, with net cash as % of assets to normalize across players of different scale. Mega-caps (compared with individual players) are seen as having greater resources to accumulate compute, fund multi-modal exploration and push faster product distribution.

Valuation multiples: For independents, P/ARR 2026E multiples are used as they are not yet profitable and P/ARR is a more comparable measure of business scale and future monetization potential, and for mega-caps the P/E 2026E is used.

Appendix

China's Key Players at a Glance:�Mega-caps

China's Key Players at a Glance: Key Independent Players

Much more, including the full assessment of multi-generational models, as well as upside and downside risks, in the full Goldman note available to pro subscribers.

Tyler Durden Sun, 07/19/2026 - 22:45
Close

Mon, 20 Jul 2026 02:10:00 +0000

'Going Dark': The High-Tech Tricks Behind Illegal Chinese Fishing In The Pacific

'Going Dark': The High-Tech Tricks Behind Illegal Chinese Fishing In The Pacific

Author Read more.....

'Going Dark': The High-Tech Tricks Behind Illegal Chinese Fishing In The Pacific
Authored by AAP via The Epoch Times,

The interception of a Chinese vessel accused of illegal fishing highlights the technical lengths many use to remain undetected across millions of square kilometres of open ocean.

Tuvalu Police operating on a Sea Shepherd patrol boat boarded and arrested a Chinese longliner, the Lu Rong Yuan Yu 138, on July 9 for allegedly illegally operating within the small South Pacific nation's waters.

In a supplied image, fish found on board Chinese longlining vessel, the Lu Rong Yuan Yu 138, at the time it was detained for alleged illegal fishing in NA, Tuvalu on July 9, 2026 PR Image/Supplied by Sea Shepard

The vessel was allegedly running with its navigation light turned off and transmitting false data showing it to be in different area - an increasingly common practice known as "spoofing."

After boarding the ship, officers found it did not hold a valid permit to fish the area, detaining the vessel and escorting it to the Port of Funafuti for further investigation.

Sea Shepherd Australia Chair Peter Hammarstedt said illegal fishers were increasingly "going dark" by switching off their automatic identification systems, or using them to broadcast false information.

"It's very common practice, especially in the Pacific Ocean, for Chinese fishing vessels to transmit false positions," Hammarstedt told AAP.

"If you go onto some vessel tracking software websites, you'll see some Chinese fishing vessels appear on land, like in the Mongolian desert, for example.

"Obviously the fishing vessel isn't there, but is operating elsewhere and simply inputting false data to mislead investigators."

In the case of the Lu Rong Yuan Yu 138, it was allegedly transmitting a location that put it squarely on the equator at 0 degrees, raising the suspicions of investigators.

Sea Shepherd has partnered with Tuvalu Police for the past two years, since the nation's only operational patrol boat was damaged in a cyclone.

The small island nation monitors a marine domain covering roughly 750,000 square kilometres, with limited assets to keep an eye on such a large area, meaning most illegal fishing operations go undetected.

"The fishing vessels that do this kind of behaviour operate under the expectation that they're just not going to get caught," Mr Hammarstedt said.

Australia and Pacific nations have recently ramped up efforts to clamp down on illegal fishing under the Pacific Islands Forum Fisheries Agency Fishing Authorisation.

"The Tuvaluans live by the sea, with the sea, and have done so for hundreds and hundreds of years," Hammarstedt said.

"When I see illegal fishing on an industrial scale happening in the Tuvaluan waters, I see this as theft from the Tuvaluans."

Tyler Durden Sun, 07/19/2026 - 22:10
Close

Mon, 20 Jul 2026 01:35:00 +0000

India's Nuclear Push Further Strains The Massive Uranium Supply Gap

India's Nuclear Push Further Strains The Massive Uranium Supply Gap

Prime Minister Modi secured another major uranium supply line Read more.....

India's Nuclear Push Further Strains The Massive Uranium Supply Gap
Prime Minister Modi secured another major uranium supply line. During his visit to Australia this month, the two sides finalized the administrative arrangements needed to move Australian uranium to India under the previous 2014 nuclear cooperation agreement. The fuel must stay under IAEA safeguards for exclusively peaceful use.�

Australia holds roughly 28% of global uranium reserves. New Delhi has eyed those reserves for years, and now the yellowcake spigot can open.

This follows the March deal with Canada's Cameco. That contract covers nearly 22 million pounds of U3O8 from 2027 through 2035 at market-related pricing, worth around C$2.6 billion.�

India already buys from Russia and Uzbekistan, and holds some domestic production. The message is consistent: a country targeting 100 GW of nuclear capacity by 2047 cannot rely on thin indigenous resources alone.�

Current operable capacity sits near 8 GW. The gap requires sustained annual additions measured in gigawatts, not megawatts.

Cumulative net deficits keep expanding as reactor builds in China and Russia outpace primary supply response. Goldman models have only recently started taking into account the significant expansion in the SMR space as well, resulting in some relatively alarming supply gaps in the years ahead.

The Cameco agreement already reflected this reality. CEO Tim Gitzel noted that sovereign buyers are locking up volumes in a window where available supply grows more uncertain. Kazakhstan has also moved to build its own strategic reserve.

Couple these data points with Goldman's estimate of the ongoing lack of long-term uranium supply contracting by the existing global fleet�

We've been pointing to the uranium upside trade for several years now. We laid out a longer form of our thesis for the anticipated continued rise of the price of uranium just a few months ago. The idea is incredibly straightforward. There's not enough mines, and the demand is exploding faster than people can keep track of.

Tyler Durden Sun, 07/19/2026 - 21:35
Close

Mon, 20 Jul 2026 01:00:00 +0000

The Data-Center Revolt Goes National: Tea Party Veteran Leads 142 Rallies Across 42 States

The Data-Center Revolt Goes National: Tea Party Veteran Leads 142 Rallies Across 42 States

The backlash against the AI data-center build-out - which we've been tracking since it was a Read more.....

The Data-Center Revolt Goes National: Tea Party Veteran Leads 142 Rallies Across 42 States

The backlash against the AI data-center build-out - which we've been tracking since it was a smattering of county fights across 28 states - staged its first coordinated day of action on Saturday: 142 protests across 42 states, from Wasilla, Alaska to Naples, Florida, organized by Humans First, the nonprofit that Tea Party veteran Amy Kremer co-founded and chairs. The crowds spanned both sides of the aisle...uniting a MAGA stalwart in a 'faith, family, freedom' T-shirt in New Jersey, a first-time activist in Texas, and a left-leaning organizer in California's Imperial Valley.

The fight looked like Kenilworth, New Jersey. Residents of the 8,500-person borough gathered outside the municipal court at mid-morning with drums, plastic horns, and sidewalk chalk to protest the $1.8 billion CoreWeave AI data center their planning board approved in May 2025 on the former Merck campus - a project that has since drawn more than 12,000 petition signatures against it, several thousand more names than the town has people. The woman in the "faith, family, freedom" T-shirt marched beside neighbors holding "Build community, not data centers" signs. When heavy rain arrived later in the day, they pulled on ponchos, shared umbrellas, and kept marching. One sign, caught by Business Insider's photographer on the scene: "You think this is pressure? Wait 'til there's no water pressure."

Texas, the country's hottest data-center market, hosted the most rallies - 18 - with Georgia at 11, California at eight, and Pennsylvania, Florida, and Indiana at seven apiece. In Imperial Valley, where a proposed facility could pull 260 million gallons a year from the Colorado River, Ivan DelSol, 54, told Reuters that around 50 people turned out in 100-degree heat. "It's dystopian that you would use this much fresh water for AI," he said. Organizers released no headcounts; turnout ran below expectations in some rural areas and in Atlanta, where about a dozen showed - most of them, a volunteer there said, driving in from the smaller Georgia towns where the biggest data centers are going up.

Kremer is a founding figure of the Tea Party movement who went on to found Women for Trump, and an organizer of the January 6, 2021 rally that preceded the Capitol riot (she neither planned nor took part in the riot itself). She has spent months calling data centers the defining fight of her lifetime, warning the technology could threaten humanity itself, and she is open about running the old playbook: grassroots pressure, town by town, aimed at both parties.

Today was incredible!

142 protests. 42 states. One message: ONE NATION UNITED AGAINST DATA CENTERS!

Communities across America showed up today for a National Day of Protest Against AI Data Centers to say that We The People deserve a voice before massive data centers are� pic.twitter.com/TwE7oRjR6V
� Kylie Jane Kremer (@KylieJaneKremer) July 18, 2026

Her crowds bear that out. One of Saturday's Texas rallies, in Tyler, was organized by Eva Cardona, a 31-year-old self-described political nomad and first-time activist who told Reuters she wanted something more hands-on than posting on Facebook; about a dozen people came. And in Imperial Valley, the man who helped lead the rally leans left. The polling explains why a coalition that broad holds. A June Reuters/Ipsos survey found only 14 percent of Americans would support a data center in their own community. Gallup polling fielded in March found 71 percent oppose building an AI data center in their area - 48 percent strongly - a worse number than a local nuclear plant gets. And Morgan Stanley told clients in a July 14 note that support for local data-center bans runs strongest among Republican, higher-income, and urban voters, while Morning Consult's national tracker crossed a line of its own in May: "stop building" (about 45 percent) overtook "keep building while expanding energy supply" (about 38 percent) for the first time since last October.

Several democratic candidates, including candidate for Texas Governor Gina Hinajosa, stopped in Lubbock to discuss campaign topics including data centers. A data center protest was also held to collect signatures to pause building hyperscale data centers in the Lubbock area. pic.twitter.com/wh3mdGNcu4
� KAMC News (@KAMCNews) July 19, 2026

For all the movement's reputation, Kremer's demands stop well short of a shutdown. She opposes a national moratorium and statewide moratoriums alike, telling Business Insider that each community should choose what gets built inside it, and that too many of those choices are made behind closed doors. Humans First's platform runs to transparent approval processes, environmental review before permits are granted, union construction jobs, and binding developer commitments of the kind lawyers call community benefits agreements. She has aimed as much fire at her own side, accusing Republicans of giving Big Tech a free pass and predicting the industry will cozy up to Democrats the moment the majority flips. The fix, she argues, belongs to Congress.

The organization is a narrower thing than the crowds it convened. Humans First announced in April that its non-conservative team members would spin off into a separate group, with Kremer promising "a topflight team of conservatives" to fight Big AI and its lobbyists. The banner over Saturday's rallies, in other words, was a conservative one - which makes the mix of people who marched beneath it all the more striking.

Official Backlash

The rallies capped a fast-moving week. On Tuesday, New York Governor Kathy Hochul signed an executive order imposing the nation's first statewide moratorium on new hyperscale data centers - an immediate pause of up to a year on state environmental permits for projects of 50 megawatts or more while regulators draft standards covering energy demand, water use, and air quality. Her office promised localities community-benefit guidance within 60 days, and Hochul will pursue repeal of the state's sales-tax exemptions for massive data centers. A tougher bill passed by the legislature, with a 20-megawatt threshold, remains unsigned on her desk; her office has called it complicated, and Hochul said the state wants to be "the first to get it right."

New York was not alone. Virginia's new tax on data-center electricity - 1.1 cents per kilowatt-hour - took effect July 1. Pennsylvania's House passed a ban on non-disclosure agreements in data-center deals by a 171-31 vote, and separately voted 197-5 to repeal the industry's sales-tax exemption, a break worth roughly $517 million a year by 2030. Arizona's governor signed a three-year moratorium on new data-center tax breaks in June. Legislators have filed more than 300 data-center bills this year; local pauses have passed at the county, city, and tribal level in at least 15 states.

Meanwhile In China

Nothing comparable is happening on the other side of the Pacific. Two days before the marches, Beijing-based Moonshot AI released Kimi K3, a 2.8-trillion-parameter system billed as the largest open-weight model ever built and claimed to perform level with America's best frontier models. Bloomberg has reported that Beijing plans to spend roughly $295 billion over five years on a nationwide network of AI computing hubs - and none of it will face a zoning board. Under the state's "Eastern Data, Western Computing" program, the buildout is steered into the arid, sparsely populated west; provincial governments compete to attract data centers with tax holidays, cheap land, and compute vouchers, and the state absorbs up to half of operators' energy costs, so the strain never shows up on a household bill. No Chinese county has passed a moratorium, because no Chinese county gets a vote.

The financial toll is no longer hypothetical. Third-party trackers cited by Morgan Stanley in its July 14 note put the value of cancelled or delayed projects at roughly $156 billion in 2025 and another $130 billion in the first quarter of 2026 alone - about $286 billion in all - set against the bank's own estimate of $877 billion in AI capital spending this year. The underlying quarterly count comes from Data Center Watch, a tracker run by 10a Labs, an intelligence firm whose client list includes AI companies, which logged at least 75 projects blocked or delayed from January through March - matching in one quarter the number of projects derailed in all of 2025 - as active opposition groups more than doubled from 396 to 833 and spread to 49 states.

Saturday itself stayed peaceful - chalk, chants, drums, umbrellas - though the wider fight has had harder edges: developers of the Piedmont transmission line into Northern Virginia's data-center corridor asked a federal court last summer for U.S. Marshals to escort survey crews after landowners threatened workers, an episode we covered at the time.

The industry answered forcefully. The Data Center Coalition warned that New York's moratorium tells investors the state is "closed for business" and will push jobs and tax revenue to neighboring states. Seven major AI and cloud companies - Amazon, Google, Meta, Microsoft, OpenAI, Oracle, and xAI - point to the Ratepayer Protection Pledge they signed at the White House in March, a voluntary commitment to cover the grid costs their facilities create. And White House AI czar David Sacks went after Hochul's case point by point on the All-In podcast, calling data centers "the scapegoat for all of the angst that people have about AI." His answer to the utility-bill complaint: let developers build their own generation behind the meter instead of competing with households for grid power.

David Sacks: Everything Kathy Hochul Said About Datacenters Was a False Accusation@DavidSacks: �These data centers become the scapegoat for all of the angst that people have about AI.

Everything she's saying there is a false accusation on datacenters. So let's just go one by� pic.twitter.com/gAxesw3fRD
� The All-In Podcast (@theallinpod) July 18, 2026

Some industry allies go further, pointing to OpenAI's June disclosure that it banned a China-linked network using ChatGPT to mass-produce comics and comments blaming data centers for rising power bills. OpenAI itself found the operation gained almost no authentic traction - and the crowds in Kenilworth and Imperial County were unambiguously homegrown.

Kremer spent Saturday evening thanking volunteers and looking past the weekend. "America is not for sale, and our communities are not collateral," she wrote on X. She expects data centers on the ballot in November, and again in 2028. The midterms are less than four months away.

Tyler Durden Sun, 07/19/2026 - 21:00
Close

Mon, 20 Jul 2026 01:00:00 +0000

478 Individuals Charged In Border-Related Enforcements In Texas, Arizona

478 Individuals Charged In Border-Related Enforcements In Texas, Arizona

Authored by Naveen Athr Read more.....

478 Individuals Charged In Border-Related Enforcements In Texas, Arizona
Authored by Naveen Athrappully via The Epoch Times,

A total of 478 individuals have been charged across Texas and Arizona over the past week in cases involving border and immigration-related crimes.

A United States Border Patrol agent enters through the United States border wall outside of San Diego, on Jan. 20, 2026. John Fredricks/The Epoch Times

In Texas, 199 people, including 175 illegal immigrants, were charged in 194 cases between July 10 and 16, according to a July 17 statement from the Department of Justice (DOJ). Authorities filed 146 criminal complaints in the Southern District of Texas for felony reentry after prior removal. Illegal entry charges were filed against 17 individuals.

According to the DOJ, most of the arrested illegal immigrants have prior felonies, such as violent crimes, immigration crimes, and narcotics. Twenty-eight people were charged with human smuggling, including 12 illegal immigrants. Charges related to firearms have also been leveled against some of the apprehended people. One of the arrested individuals has a prior conviction for online solicitation of a minor.

In Arizona, 279 individuals were charged during the July 11-17 week following immigration enforcement operations, according to another DOJ statement on July 17.

Authorities filed 56 cases in the District of Arizona in which illegal immigrants were accused of reentering the United States unlawfully. Officials charged 206 illegal immigrants with illegally entering the country. Cases were filed against 17 individuals for allegedly smuggling foreign nationals into Arizona.

The 478 arrested individuals were apprehended under Operation Take Back America, established by the then-deputy attorney general Todd Blanche through a March 2025 memorandum.

The operation seeks to repel the entry of illegal immigrants into the United States, achieve the total elimination of cartels and transnational criminal organizations, and set up Homeland Security task forces to dismantle cross-border human smuggling and trafficking networks.

In California, 111 border-related cases were filed this week, the DOJ said in a July 17 statement. Charges filed at the Southern District of California include reentering the United States after deportation, importing controlled substances, and bringing in illegal immigrants for financial gain.

Among the arrested are a Mexican national with three prior felony immigration convictions, and a U.S. citizen who tried to traffic 131 pounds of cocaine when applying for entering the United States.

ICE Enforcement

The Trump administration is facing opposition from certain jurisdictions over immigration-related arrests.

In April, the Philadelphia City Council voted unanimously to pass the ICE OUT legislation package, which placed restrictions on Immigration and Customs Enforcement (ICE) activities in the city, according to a council statement on April 15.

The legislation bans the city from collaborating or sharing data with ICE, restricts ICE activity in city-owned spaces, and makes it illegal for ICE and other law enforcement agents to hide their identities with face masks and unmarked vehicles.

The committee's vote on the ICE OUT package "affirms that we will stand with our immigrant communities and refuse to turn our backs on those who call this city home," Councilmember Rue Landau said in a statement at the time.

A federal judge temporarily blocked the ban on face masks and unmarked vehicles on July 2 after the federal government requested a preliminary injunction. The prohibition was set to come into effect on July 7.

Judge Chad Kenney of the U.S. District Court for the Eastern District of Pennsylvania, who issued the ruling, said the city's law "attempted to sidestep the Constitution's clear mandate."

Meanwhile, ICE said in a July 1 statement that the Secure America Act, signed into law by President Donald Trump in June, provides the agency with $350 million in additional funding that would enable the arrests of more criminal illegal immigrants hiding in sanctuary jurisdictions.

Acting ICE director David J. Venturella said in the statement that sanctuary jurisdictions refuse to honor ICE detainers or coordinate to safely transfer illegal immigrants from local custody to federal authorities.

"As a result, criminal illegal aliens who could have been transferred directly to ICE are instead released into our communities, forcing our officers to locate and arrest them in neighborhoods, businesses, and other public locations," Venturella said.

"The funding from this legislation will help ICE increase its capacity to monitor releases and arrest removable criminal aliens when sanctuary jurisdictions refuse to cooperate with federal immigration authorities."

Tyler Durden Sun, 07/19/2026 - 21:00
Close

Mon, 20 Jul 2026 00:25:00 +0000

China's Moonshot Eyes IPO After Sparking Another "DeepSeek Moment" For US AI Stocks

China's Moonshot Eyes IPO After Sparking Another "DeepSeek Moment" For US AI Stocks

Chinese startup Moonshot AI sparked another Read more.....

China's Moonshot Eyes IPO After Sparking Another "DeepSeek Moment" For US AI Stocks
Chinese startup Moonshot AI sparked another DeepSeek-like moment for US AI stocks and hyperscalers last week. David Sacks, the former White House AI and cryptocurrency czar, warned that this is "very concerning."

Moonshot AI debuted Kimi K3, which ranks number three on the Artificial Analysis Intelligence Index, trailing only OpenAI's GPT-5.6 Sol and Anthropic's Fable 5. This only shows the continued trend of open-weight models toward the AI frontier.

Kimi K3 has emerged as a top challenger across reasoning and coding workloads, highlighting the rapid pace of innovation and the increasing competitive pressure on models in terms of capabilities and downward trend in pricing. This only sparked weakness in sentiment toward AI US stocks and hyperscalers last week due to their massive capex binges to build out data centers.

Related:

Goldman Warns Massive Hyperscaler Bond Issuance Means "Risk Skewed To Downside"

It serves as a reminder for companies amid the 'tokenmaxxing' woes that there are lower-cost alternatives for end users besides expensive frontier Anthropic and OpenAI.

Now, Moonshot AI is leveraging last week's momentum and preparing for an initial public offering in Hong Kong within six months. Kimi K3 has strengthened investor confidence, positioning the startup to enter public markets.

Bloomberg reports that Moonshot is in the process of concluding a funding round that values the three-year-old startup at $30 billion.

Annual revenues topped $300 million in June, up from $200 million in April, supported by chatbot subscriptions and enterprise services.

The startup has held talks with CICC and Goldman Sachs about the offering and is dismantling its red-chip structure to facilitate an overseas listing.

Meanwhile...�

We�ve identified industrial-scale distillation attacks on our models by DeepSeek, Moonshot AI, and MiniMax.

These labs created over 24,000 fraudulent accounts and generated over 16 million exchanges with Claude, extracting its capabilities to train and improve their own models.
� Anthropic (@AnthropicAI) February 23, 2026

Here is the latest institutional commentary on Moonshot AI:

Jefferies analyst Roger Song

Moonnshot AI triggered another Deepseek-like moment + semi/ memory lost momentum, both potentially benefiting biopharma rotation. While still waiting for FDA leadership renewal, we started to see more potential policy reversals (while maintaining good ones, nice to see CRL release back for now) along with swift industry reactions. Open clinical trial prediction betting - will that be a distraction/ noise or another actionable datapoint for real investment? We shall all see.

Wolfe Research analyst Alex Zukin

China's Moonshot Unveils Kimi K3, Autonomously Designed Functional Chip in 48 Hours - Moonshot released Kimi K3, a 2.8 trillion parameter Mixture-of-Experts model with a 1-million-token context window and native text, image, and video support, positioned as the first open 3T-class frontier model with full weights releasing July 27, 2026 and API pricing at $3/$15 per million input/output tokens. The model autonomously designed a functional 4mm� chip in a single 48-hour run with zero human intervention, handling design, optimization, verification, and simulation, achieving over 8,700 tokens/second decoding throughput, while also building a custom GPU compiler called MiniTriton from scratch that reportedly matches parts of Nvidia's official Triton compiler. Kimi K3 scored 88.3 on Terminal-Bench 2.1, trailing but competitive with OpenAI's GPT-5.6 Sol, while lagging flagship closed models like Anthropic's Claude Fable 5 in broader evaluations, with the model available today via the Kimi app, Kimi Code, and OpenRouter

Canaccord Genuity analyst Martin Roberge

Meanwhile, sentiment toward AI stocks and hyperscalers' capex binge keeps weakening and today's release of a new AI model by Chinese company Moonshot, capable of competing with AI models from OpenAI and Anthropic, serves as a reminder that there are lower-cost alternatives for end users.

The race among AI startups to go public has China-based DeepSeek targeting a 2027 IPO, while US rival Anthropic could list as early as late 2026 and OpenAI increasingly appears headed for a 2027 debut. Both US firms have confidentially submitted IPO paperwork.

Tyler Durden Sun, 07/19/2026 - 20:25
Close

Sun, 19 Jul 2026 23:50:00 +0000

Taiwan's President Ratchets Anti-China Rhetoric In Party Speech: Must Resist 'Red Terror' Of Beijing

Taiwan's President Ratchets Anti-China Rhetoric In Party Speech: Must Resist 'Red Terror' Of Beijing

President Lai Ching-te said on Sunday that Taiwan must renew efforts to protect its democracy and never become part of China. He ur Read more.....

Taiwan's President Ratchets Anti-China Rhetoric In Party Speech: Must Resist 'Red Terror' Of Beijing

President Lai Ching-te said on Sunday that Taiwan must renew efforts to protect its democracy and never become part of China. He urged members of his Democratic Progressive Party (DPP) to oppose the "red terror" coming ?from Beijing.

The ruling DPP champions Taiwan's separate ?identity from China, and is working to combat what�Lai called�China's "legal warfare".

"I also expect comrades within ?the party to stand on the front lines, unite as one, and jointly oppose the ?threat posed by China's 'red terror' ?to Taiwanese society," Lai added, speaking in Taiwanese, rather than the main language of government, Mandarin.

"We must work together ?to protect our democratic and free way of life, and absolutely never allow 'democratic Taiwan' to turn back and become 'China's Taiwan'," he said.

"Regardless of ethnic group, regardless of who came earlier or later, anyone who identifies with Taiwan is a master of ?the country. Taiwan's future must be decided jointly by the 23 million people of Taiwan," he said.

One establishment US publication has held up a prime example of Beijing's "legal warfare" against an autonomous Taiwan in the following:

On March 12, China�s legislature�adopted�the Law on Promoting Ethnic Unity and Progress (Chinese;�English translation), a sweeping new statute that codifies Beijing�s approach toward China�s 56 officially recognized ethnic groups.�Substantively, the law enshrines a decades-long shift towards aggressive assimilationist policies.�Structurally, it reflects a deepening merger of Party ideology and state law that is becoming increasingly prevalent under Xi Jinping.

This new law is the culmination of a policy trajectory that has been building for over a decade, dating back to the�2014 Central Ethnic Work Conference.�Under Xi, Beijing is steering away from the post-1949 legal framework of nominal ethnic autonomy (albeit under tight Party control) imported from the Soviet Union.�In its place, officials have steadily been pivoting towards what scholars have termed �second-generation ethnic policies��an aggressive assimilationist approach that emphasizes a common Chinese national identity over accommodation of ethnic differences. Provincial and municipal authorities across China have enacted a wave of local �ethnic unity and progress� regulations in recent years, such as those in�Xinjiang�(2015) or�Inner Mongolia�(2021). The new national legislation elevates this approach to the level of a national statute governing all of China.

The new law�s core concept is captured in the term�zhulao�� to �forge� or �cast� metal � and its instruction that �forging the communal consciousness of the Chinese nation� is core to the Party�s ethnic policies.� As James Leibold has�pointed out, this phrasing reflects a hardening of Beijing�s political line under Xi Jinping � explicitly written into the Party�s�Charter at the 19th�Party Congress in 2017 � aimed at �melting� subnational and ethnic identities into a shared collective one.

Meanwhile, on a global stage Beijing has continued to present itself as the only peace guarantor and as a�force for stability�and is seeking�'Taiwan's willing participation' - at a moment the Middle East is on fire�largely as a result of American policy and quickness to result to force and surprise attacks.

President Donald Trump suggests the United States might not come to Taiwan�s aid if China attacks, arguing that Taiwan �took� America�s advanced semiconductor business and should therefore pay more for its own protection. � @atrupar (July 2023)pic.twitter.com/WNMUZe6IUf
� The Intellectualist (@highbrow_nobrow) July 19, 2026

And yet, President Trump has of late publicly touted his personal relationship with Chinese President Xi Jinping as "amazing". Planned weapons deal with Taiwan have been indefinitely put on hold as Washington tries to repair relations with Beijing.

Tyler Durden Sun, 07/19/2026 - 19:50
Close

Sun, 19 Jul 2026 22:40:00 +0000

Judge Strikes Down Race-Based Provision In Biden-Era Internet Access Grant Program

Judge Strikes Down Race-Based Provision In Biden-Era Internet Access Grant Program

Au Read more.....

Judge Strikes Down Race-Based Provision In Biden-Era Internet Access Grant Program
Authored by Aldgra Fredly via The Epoch Times,

A federal judge ruled on July 15 that a race-based provision of the Digital Equity Act, signed by President Joe Biden in 2021 to close digital gaps, was unconstitutional.

A judge's gavel rests on top of a desk in a courtroom in Miami, Fla., on Feb. 3, 2009. Joe Raedle/Getty Images

The Digital Equity Act was part of Biden's Infrastructure Investment and Jobs Act, which appropriated $2.75 billion to the National Telecommunications and Information Administration (NTIA) to establish grant programs to expand high-speed internet access for minority groups and communities in rural areas.

After taking office for a second term last year, President Donald Trump halted the competitive grant program authorized under the Digital Equity Act, saying it was unconstitutional because it allocated federal funding based on race.

The National Digital Inclusion Alliance, a recipient of the competitive grant program, later filed a lawsuit in October 2025 seeking to reinstate the program.

In a 35-page order, U.S. District Judge John Bates ruled that the Digital Equity Act's provision authorizing the use of race in awarding federal funds was unconstitutional, citing the Supreme Court's 2023 ruling that struck down race-based preferences in higher education admissions.

Bates said that while the Digital Equity Act aims to address the digital divide among minority groups and other covered populations, the Supreme Court precedent showed that remedying general social disparities alone does not justify the use of race in government action.

"Addressing that gap is a laudable goal, but the Supreme Court has admonished that ameliorating general societal inequalities - as opposed to specific instances of past discrimination - 'does not constitute a compelling interest that justifies race-based state action,'" the judge stated.

"Otherwise, Congress could deploy racial classifications when confronted with any situation of an uneven resource distribution."

Bates said the grant program could be reinstated without the race-based provision, and the government had committed to restoring it upon a judicial determination that the provision was unconstitutional.

Trump welcomed the ruling in a Truth Social post, calling it a "big win" for the American people.

"The so-called 'Digital Equity Act,' a Biden DEI law, was ruled exactly what I said it was last year - A RACIST and UNCONSTITUTIONAL giveaway that never should have become Law," he wrote.

The decision to end the Digital Equity Act comes amid the Trump administration's efforts to eliminate diversity, equity, and inclusion (DEI) programs from federal agencies and government initiatives.

Trump stated in a Jan. 20, 2025, executive order that the previous administration had forced "illegal and immoral discrimination programs" across virtually "all aspects of the federal government" through DEI initiatives.

Christopher Mitchell, director of the Community Broadband Networks Initiative at the Institute for Local Self-Reliance, credited the National Digital Inclusion Alliance with helping to secure the program's restoration.

"Yesterday's ruling on the Digital Equity Competitive Grant Program is, on balance, a victory," Mitchell said in a statement. "The only real question now is how quickly NTIA moves to actually implement it."

The National Digital Inclusion Alliance did not return a request for comment by publication time.

Tyler Durden Sun, 07/19/2026 - 18:40
Close

Sun, 19 Jul 2026 21:30:00 +0000

NBA Boss Calls Caitlin Clark A "Political Football" Over Racism Controversy

NBA Boss Calls Caitlin Clark A "Political Football" Over Racism Controversy

The WNBA has been the butt of jokes for decades.� The women's league is a perpetual money pit, with annual losses of around $40 million per year, all subsid Read more.....

NBA Boss Calls Caitlin Clark A "Political Football" Over Racism Controversy

The WNBA has been the butt of jokes for decades.� The women's league is a perpetual money pit, with annual losses of around $40 million per year, all subsidized by the men's league.� The level of play is often tedious, the pace is much slower and the athleticism is lacking compared to the men's game.�

It's the reason why the WNBA has far less viewers, makes no profit and why the women athletes are paid far less.� It's not sexism ad many of the players claim, it's just basic math.

One would think that any player that doubles (in some cases triples) the viewership of league games by her mere presence would be widely celebrated by all the people involved including the commission, but sadly, this has not been the case with Caitlin Clark.� Clark's expertise and superior ability on the court has brought new life to a floundering sport; the problem is, she's relatively new to the league and, she's white.�

This combination triggered many of the minority players, leading to public back-biting on sports talk shows and flagrant fouls on the court.� The audience and social media started to take notice.� In fact, one could argue that it was all the clips of jealous WNBA athletes attacking Clark on the court that made her more popular than ever.� Not to mention, it seems as if most of the fouls are committed by black players.�

The latest flagrant foul involved Phoenix�Mercury forward Alyssa Thomas, who appeared to knee Clark in the groin area and then press her closed fist into Clark's throat while she was on the ground.� No foul was called.�

??Hold up�A closer look at the dirty play by Alyssa Thomas last night w/ a new angle � Wait for the freeze + zoom on each.

- Knee to the groin
- Pinches her underarm
- Fist to the throat
- Attempt to stomp on her (she moved)

Why do @WNBA players hate Caitlin Clark so much? ?? pic.twitter.com/83JZSChdEu
� HeroOfTheDay (@Hero_OfThe_Day) June 25, 2026

Clark suffered a back injury from the incident and has to sit out multiple games.� This is a regular occurrence with Clark; she stands near the top of the list for most fouls and flagrant fouls in the WNBA.� In many cases involving flagrant fouls, referees do not make the call, leading to anger and social media backlash among fans.�

Some of the dirtiest tricks in basketball are on display regularly in WNBA games, and the attacks on Clark look practiced and organized.

� �

It's not always black players that go for the cheap shots, but there is a clear pattern emerging.� It could be envy over Clark's far larger fan following despite being a newer player (female cattiness), but suspicions are hanging in the air that her treatment is largely race motivated.� By far, Clark's most vocal critics in sports media are black.

The WNBA and the NBA are not happy with the political undertones of the Clark controversy.� NBA Commissioner Adam Silver recently called Clark a "political football" in interviews as he attempted to dismiss the WNBA's numerous problems with officiating fouls.� He seems to outright dismiss any concerns about race-based targeting of Clark.

Speaking as part of a panel at an event in New York, Silver said the debates surrounding Clark had become about broader political and cultural issues in the United States rather than basketball alone.� He specifically referenced the foul by Alyssa Thomas that led to Clark's injuries.��

"That particular incident is not about whether a foul should have been called at the time of the game or whether that was ultimately a flagrant non-review....I've come to know Caitlin really well. She's an incredible player and also an incredible person.� "And she wants to focus on being the best player she can. And she's become a bit of a political football in this country, and I think it's incredibly unfair to her."� ��

Speaking of unfair, Clark was snubbed by the WNBA in a 30th Anniversary commemorative poster (she was strangely absent from the roster), and she was snubbed by the 2024 Olympic "Dream Team" despite being a top star and #1 draft pick.�

Emmanuel Acho, a former NFL player and sports analyst (who is also coincidentally black), asserted on the Speakeasy podcast in June that Clark has "become a distraction" and that the WNBA would be better off without her.� He argued that Clark got audiences to watch the games, and now the WNBA can throw her away.�

????WORST TAKE OF ALL-TIME????

Ex-#NFL LB Emmanuel Acho says the WNBA would be better off without Caitlin Clark because she is a distraction.

�The WNBA would be better without Caitlin Clark, because she is a bigger distraction than she is an additive� pic.twitter.com/MVwHoWRS8u
� MLFootball (@MLFootball) June 29, 2026

They hate Clark and her popularity with a passion, but they like the idea of stealing her audience for themselves.� It's rather pathetic, but we're talking about the WNBA.� They have lived in the gutter for so long they can't imagine climbing out of it.� Of course, Acho is delusional if he thinks all the "eyes" Clark brings to the WNBA won't leave with her if she exits the league.� �

The bottom line is this: Clark puts butts in seats.� No other WNBA player comes close.� Yet, the league and many of its commentators treat her like a pariah because they can't make their crow's nest of minority players behave.� They are the cause of the controversy.� They made Clark into a "political football" by remaining apathetic and dismissive.��

Tyler Durden Sun, 07/19/2026 - 17:30
Close

Home     About Us     Business Services     Business Exchange    Executive Network    Recruit Executives    Join Us    Contact Us
Senior Executives     Business Entrepreneurs      Investors & Ventures      FundEX      Interim & Permanent Management
Special Programmes  Login

Dotsquares