Connexion
This by Scott Ritter is an essay that is complimentary to his group of interesting dinner companions.
This is not what I want to highlight, though, but read the piece here:
What I would like to highlight, is this:
The conversation then turned to the reality of the modern battlefield. I asked Apti to comment on President Trump’s assertion that Russia was losing the war. Apti’s response was surprisingly diplomatic, given the absurdity of the assertion. The Hero of the Russian Federation noted that while the situation on the battlefield in the Kursk region might have been somewhat fluid this time last year, today the reality was far different—the front had stabilized, and the Ukrainian forces had been driven out of Russian territory. Moreover, Apti noted, never once—even in the most difficult of times—had the Akhmat Special Forces ever surrendered a position they held to the Ukrainians. To the contrary, the Akhmat Special Forces have been systemically advancing across the breadth of their assigned area of responsibility, whether in Lugansk, Kursk or Belgorod.
When Apti and I first met in person, back in May 2023, we had a very frank and detailed discussion about the art of war, and how the progression of technology had made past practices—including those perfected by the US military in the 1980’s, when I was learning the art of war—obsolete on the modern battlefield. Apti was familiar with my experience as an artillery forward observer and forward air controller, as well as my background in combat intelligence collection. He pointed at me and reiterated the point he had first made back then. “Your experience is no longer viable,” he said. “It has been taken over by drones. It is a completely new kind of war being fought today with Ukraine, one the West is not prepared for.”
We then entered an extended group conversation about drone warfare, and how the Akhmat Special Forces had evolved in this area. “We have the best drone unit in the Russian army,” Apti declared, “which means the best in the world. Our drones are thoroughly integrated into every aspect of our operations.” Apti explained, in general terms, how modern drone warfare was practiced. “Objective control” over the battlefield was maintained on a constant basis, with highly specialized reconnaissance and surveillance drones keeping watch. These drones were supported by mid-level reconnaissance drones, which could respond to any cues provided by the overwatch drones to provide increased resolution and fidelity to any observed activity. And then there was the fire support, provided by a multitude of different drone models which could respond to different target sets at different ranges under varying circumstances. Suffice it to say that hundreds of drones were available to the Akhmat Special Forces daily, and effective and efficient supply lines ensured that the Akhmat Special Forces never lacked resources when it came to drones.
Apti was very proud of the men who oversaw his drone operations, praising not only their tactical skills but also their innovation and intellect when it came to the technologies associated with drone operation. He particularly singled out his electronic warfare operators, whom he claimed were the best in the business. One of the reasons for this, Apti maintained, was the quick learning cycle of the Russian military, and the Akhmat Special Forces in particular. Innovation was driven by the needs and experiences of the frontline fighters, and the Russian military and Ministry of Defense were organized and conditioned to turn battlefield innovation into Russian doctrine and procedures, backed by the proper technology and other resources.
The sheer complexity of drone warfare is staggering. The need for appropriate technology to support this kind of warfare is an unrelenting demand in today’s world. Here, we can see the necessity for using AI, as the number of decisions that need to be made quickly exceeds the ability of a human. Drones control drones that control drones: so this is a new stack of knowledge being constructed. It happens to fit thematically with a recent conversation with a friend.
Massive drone warfare ‘controlled’ by AI. This needs to be taken very seriously as there are looming threats from the US modern technology complex, dreaming of taking Taiwan with no boots on the ground, no loss of US men, and in their demented minds, such an easy win or so say the war games. Rationality is not a strong suit of the US, notes my friend.
It was only about a year ago that I asked a professional on weapons and warfare to explain to me the role that drone warfare plays on the modern battlefield. The good education that I received then is now partially obsolete within just a year. It is also not much more than a year since Nat South explained to us in a few well-constructed essays how drones work, how the Russians use them, and where the gaps lie, focusing specifically on maritime unmanned boats: From Nat South at The Lookout – Z, on Drones.
And then we have the lone droner. We know about the proverbial lone gunman, but soon to come is the proverbial lone droner. In the current sphere of Late Stage Barbarism, we will see high-level people taken out using this method. The concept was already used with one of the attempts to take out Nicholas Maduro of Venezuela, by flying a drone right at his head at the podium while he was making a speech. It was older technology, and he ducked. These days, that simple strategy will not work.
With the fast rise of robotics in China and the incredible development of military robotics, will we soon see a war fought with robots and drones? We already have robotic Olympics. Such a war, compared with other wars, is literally a cheap war. We saw in Gaza the IDF using cheap small drones, hanging ladders on them stuffed with explosives and dropping that onto buildings and people, doubling, tripling, and quadrupling the percussive and killing power, cheaply.
AI is often written about as a threat to humanity. Technology is, of course, generally neutral. It is the application that makes it evil or good. With AI, drones controlling drones in a stack of functions, and military robotics, we have truly landed in a dystopian reality, and Brain Rot is proceeding.
There is new research indicating that Large AI Language Models, fed by trivial and stupid internet chat, literally suffer from brain rot. The research is literally called LLM Brain Rot. These models become stupid and silly, and take on the characteristics of a stupid and foolish audience. Even when changing the input later, the models repair only slightly and remain dumb. Foolishness and stupidity are indeed transmissible diseases. We’re even infecting the software around us.
AIs with brain rot, drones controlling drones controlling more drones, and robotics: It is not only a new multipolarity; the sheer lineaments of our world are being shaken up. In terms of warfare, it is now clear that the one with the most effective drone technology, AI, and techniques will gain the edge for a win.
Exploring the dystopian technofeudalism at heart of the western A.I. race
The unnamed book EMPIRE OF AI is mine, based on 300+ interviews, 7 yrs covering AI, & my time as the first reporter who got extensive access to OpenAI. I sought OpenAI’s perspective throughout. For months they said it was coming. It never did. Pre-order: https://t.co/i5rNJ45fNo. https://t.co/Drr8Gzv7vW pic.twitter.com/0aJpF9P26U
— Karen Hao (@_KarenHao) April 4, 2025
The A.I. Religious Cult: “Just like empires of the past were competing against one another, so are these tech giants. There was this aggressive race of we, the French Empire, are better than the British Empire. We, the British Empire, are better than the Dutch Empire.
And we need to continue to relentlessly race and be number one because we’re the ones that have the right civilizing mission to bring modernity and progress to all of humanity.
And where their missions might have once been about scientific exploration and transparency, they are now about profit, which has resulted in a landscape where they are just scaling at all costs.”
this is about controlling “the civilizational transformation” of the “one global village.” “Our collective attention is their Manifest Destiny.” From Maggie’s “there is no society” to now, “Culture is just what is Next on your phone — and that is A.I.”
Quis custodiet ipsos custodes?
Pavel Durov, founder of Telegram, joins Human Rights Foundation CEO and founder, Thor Halvorssen, in a rare and wide-ranging conversation on digital freedom, state censorship, and the responsibilities of tech creators in the modern world. Durov recounts how Telegram became a lifeline for protest movements from Belarus to Hong Kong and what may be around the corner for encrypted communications tools. As governments demand more control over online platforms, Durov lays out his vision for a secure, decentralized future where privacy, encryption, and free expression are non-negotiable.
Shortly after this Durov announced a joint venture with Elon Musk, to integrate the GROK AI with Telegram. Why not one of the Chinese ones?
The full version, including what happened in Tehran on several fronts – IRGC, nanotechnology, academic debates – of the column originally published on Sputnik.
By Pepe Escobar at VKontakte.
ON THE ROAD IN IRAN – The International North South Transportation Corridor (INSTC) is one of the most crucial geoeconomic/infrastructure projects of the 21st century. It unites at its core three key BRICS nations – Russia, Iran and India – branching out to the Caucasus and Central Asia. When fully operational, the INSTC will offer a full trade/connectivity corridor sanctions-free, cheaper and faster than the Suez canal to a great deal of Eurasia. The geoeconomic consequences will be staggering.
The International North–South Transport Corridor (INSTC): The New Silk Road Shaking Up Global Trade. A 7,200km sanctions-proof trade route that could replace the Suez Canal for Eurasian trade.
To re-visit Iran in these times of geopolitical trouble, relentless “maximum pressure”, red lines on uranium enrichment and bombing threats could not be more pressing – and enlightening.
These have been only a few of the highlights of the past two weeks:
In essence, Wilāya…can mean 1. regent, sovereign, lord and master; 2. patron, guardian, protector, custodian. Its plural form is awliā: those of God’s creatures who have spiritual affinity and propinquity to Him, inclusive of prophets and Imāms. The walīy is the divinely-appointed Divine Guide, Just Ruler, and Guardian-Sovereign (walīy al-amr) of the affairs of the community of true believers (the mu’minīn).
Williams explains how “Wilāya has a two-fold essence: the walīy is chosen by God because of his spiritual affinity and propinquity to Him. This preeminence gives the walīy’s soul a certain sacred luminosity (due to his sacral propinquity), which in turn draws people to him, as it responds to their need for guidance and to their appreciation of beauty and nobility of character. Thus, wilāya has a top-down function [being appointed by God due to his preeminence], and at the same time, has a bottom-up function (being drawn to and appointed by the people to lead them on their common sacred purpose of their self-surrender to God’s will (islām).” Both of these aspects (the theocratic and the democratic), as Williams stresses, must be present because, “paradoxically, wilāya is meritocratic and democratic at one and the same time.”
By then it was time to hit the road. And there was only one way to go: to explore the INSTC on the ground, from the Caspian Sea to the Persian Gulf and the Makran coast in Sistan-Baluchistan all the way to Iran-Pakistan border.
Total connectivity: highway, mosque, bazaar
By an auspicious turn of events, the old school reportage/investigation actually became the plot line of a documentary, produced in Iran, shot by an outstanding crew of 6, and to be broadcast in several parts of Eurasia, including Russia and China. Here we offer the broad strokes of our travel to the heart of the INSTC.
Tehran: (L): With serious Persian brainpower – in film making, academia and Shi’ite theology. (R): At a very serious theology, Western nihilism and sufism discussion, with, among others, Alastair Crooke and George Galloway.
We started with a series of interviews in Tehran, with Central Asia analysts and most of all Mostafa Agham, the top expert of Behineh Tarabar Azhour, a transportation and logistics firm specialized in Eurasia railway corridors. These analyses offered contrasting points of view on where the INSTC should go next and what are its main challenges.
Travel along Iran’s main artery, from Tehran to Bandar Abbas, was a must – as it will conform the trans-Iran north-to-south highway axis of the corridor. That doubles of course as a cultural and spiritual pilgrimage, which in our case featured plenty of auspicious overtones.
Isfahan: (L): On the road to Isfahan, not far from Natanz nuclear plant. (R): The mesmerizing ‘Royal’ mosque.
Isfahan: (L): Isfahan architectural magic. (R): In the most astonishing square in the world.
We arrived at fabled Isfahan past sunset, which allowed us to visit the Masjed-e Shah – or “Royal” – mosque virtually undisturbed. The Royal mosque – one of the highlights of Islamic architecture – sits on the south side of the Naghsh-e square in Isfahan, one of the most extraordinary public squares in the history of art and architecture, rivaling, and arguably surpassing San Marco in Venice.
A visit to the Isfahan bazaar is also inevitable. I was looking for an old friend who sold nomad carpets – in the end, because of slow business, he relocated to Portugal – just to find his sort of heir, young, energetic, who apart from pointing me to a spectacular, rare tribal rug from northeast Iran close to the Afghan border, gave me a crash course on the effects of sanctions and the perpetual demonization of Iran in the West (“Turkey has 40 million tourists; we have two or three”). Isfahan’s neat and extremely organized bazaar offers quality handicrafts to rival Istanbul, but there’s essentially domestic tourism, sprinkled with a few foreigners mostly from Central and South Asia and some from China.
(L): Tehran. At the IRGC Aerospace Museum; reverse-engineering an American missile. (R): Qom. At the Fatima Masumeh shrine. (leading article banner is at the Fatima Masumeh shrine at 2 in the morning)
On the way back to Tehran we learned that, being a Tuesday, the revered Haram of Fatima Masumeh, the daughter of the 7th Imam Musa, in Qom was open all night. Nothing prepares the pilgrim for an arrival at nearly two in the morning to an apotheosis of gold and crystals in the heart of Qom, Iran’s second most sacred city after Mashhad. Only a few pilgrims paying their respects, some strolling around the shrine with their families or reading the Quran. A moment of quiet illumination.
Southern Caspian Sea: (L): Fishing at the Caspian, outside Bandar Anzali. (R): On the Caspian corridor: from Astrakhan to Bandar Anzali.
Afterwards it was time to hit the Caspian, and the port of Bandar Anzali, the proverbial “international bridge” where, in theory, cargo ships from Astrakhan in the Russian Caspian, as well as other Caspian-bordering states will start arriving en masse via the INSTC. In Bandar Anzali, Iran essentially imports petrochemicals, construction materials, minerals, and iron products and exports grains (soybeans, corn, barley, wheat) and crude oil.
In Tehran, Mostafa Agham, the connectivity expert, had explained in detail that perhaps the multimodal drive of the INSTC across the Caspian may not be the best idea. The Russians prefer to build a railway bordering the western margins of the Caspian; and another possibility is to use a network of already functioning railways from southcentral Russia, across Kazakhstan all the way to Aktau, by the Caspian, and then connecting across Turkmenistan to Tehran.
It’s only via a close up on Bandar Anzali that one understands the Russian rationale. One of our cameramen, in delightful broken English, coined an instant hit: “Port no exist”. Translation: the infrastructure has not been upgraded in decades, which brings us to the devastating effects of sanctions, visible in several nodes of Iran.
That was extra evidence supporting the fact that China will have a lot of work to do as part of the 25-year strategic partnership with Iran, signed in early 2021: a comprehensive agreement between two civilization-states where energy-for-infrastructure is a central plank amidst scores of BRI-related projects.
The Persian Gulf: (L): Entering Bandar Abbas port, Iran’s key INSTC port. (R): From Bandar Abbas, the Strait of Hormuz is 39 km that way.
Break to the border!
Bandar Abbas, in the Persian Gulf (italics mine), is a completely different story. That’s Iran’s main port, and a key node of the INSTC, to be connected to Mumbai and already connected to the big ports in eastern China. We had all the hard-to-get necessary permits to explore Shahid Rajae-i Special Economic Zone, crammed with containers from shipping firms such as West Asia Express and unloading scores of Chinese container cargoes. The uber-strategic Strait of Hormuz is only 39 km to the south. A few days after our visit, Iranian President Masoud Pezeshkian went straight to the point, referring to proverbial Trump threats: “Block our oil, and we’ll block the world’s energy.” Iran can do it – in a flash; were that to happen, the collapse of the global economy is guaranteed.
Additionally, port authorities explained that the recent explosion on Shahid Rajae-i – attributed to “negligence”, still under investigation and somewhat mired in controversy – was not in the port itself, but in a storage area 10 km away.
Sistan-Baluchistan: (L): The Makran coast. The fishing port of Beris. (R): At a kapar – the Sistani version of a yurt.
From the Persian Gulf we fly to the Sea of Oman – and infrastructure problems ride again: there are only two flights a week. We arrive at a minuscule military airport outside the future superstar of the INSTC: the port of Chabahar in Sistan-Baluchistan province. Baluchis are exceedingly cool, cousins to the ones on the other side of the border, in Pakistan. In bustling Chabahar, the lineaments of a boom town are quite visible.
A long walk in the port side by side with Alireza Jahan, a logistics expert and then a conversation with Mohammad Saeid Arbabi, the Chairman of the Board and Managing Director of the Chabahar Free Trade Zone could not be more enlightening.
Jahan explains how Chabahar is essential to Iran’s East Axis, serving over 20 million people not only in huge Sistan-Baluchistan but also three other Khorasan provinces, and further on to Kerman. So Chabahar is the port for an enormous hinterland, while its competitor, Gwadar in the Arabian Sea in Pakistan, only 80 km away or so, is virtually isolated.
Jahan also explains Indian investment. Tehran is investing heavily in the infrastructure and superstructure of Chabahar port, while India is investing in equipment: the Italian cranes around the port came from India. Arbabi, at the Free Trade Zone, expands on the international profile of Chabahar, which will be an absolutely key node not only for landlocked Afghanistan but also the Central Asian “stans”.
And that brings us to the local highway saga: Chabahar to Zahedan, in the Afghan border, 632 km, already an “acceptable road”, and with a companion railway to be built within the next three years, everything 100% financed by the Iranian government.
Progress at the port is steady – slowly but surely. For the moment Chabahar receives three ships from India a month and two ships from China, plus three from the Persian Gulf. The distance from Mumbai is only 4 days, and from Shanghai, 15 days. The potential for expansion is limitless.
From Chabahar, it’s on the road bliss along the spectacular, strategic, oil-drenched, semi-desert Makran coast, bordering the immaculate Sea of Oman all the way to the Arabian Sea. History looms large: this is where Alexander the Great lost as much as 75% of his army to dehydration and starvation when he was retreating across the desert to Macedonia after his tortuous two-year invasion of India.
Sistan-Baluchistan: (L): The spectacular – virgin – Makram coast. All the way to the Iran-Pakistan border. (R): WELCOME TO THE IRAN-PAKISTAN BORDER. I finally made it – by boat! Thanks to Commander Chabad, born in Chabahar, and his little brother and second-in-command. Iran is around sunset. Pakistan is ahead of us, to the left. No CIA and MI6, no instrumentalized Salafi-jihadis disguised as Baluchi nationalists. Only the silence and beauty of the Sea of Oman meeting the Arabian Sea.
Due to a concert of economic and ecological reasons, there have been plans for quite a while to relocate the capital, Tehran, to the Makran coast. Chabahar in this case would be the ideal candidate: free port, INSTC connectivity between Central Asia and the Indian Ocean.
India – which needs to step up its geoeconomic game – has noticed it. China as well. Even though China is not formally part of the INSTC, Chinese companies are bound to invest massively in Chabahar – the de facto key node for South Eurasia integration – as part of yet another vector of the Maritime Silk Road.
I had been trying for 20 years-plus to reach the Iran-Pakistan border, from either side. It was always an impossibility due to serious security reasons. I finally made it – by boat, thanks to teenage Commander Chabad, born in Chabahar, and his little brother and second-in-command.
As we sailed away from the small fishing port of Gwat, Iran was located around sunset and Pakistan ahead of us, to the left, in what the locals call “Pakistan beach”. No CIA and MI6; no instrumentalized Salafi-jihadis disguised as Baluchi nationalists; no sanctions; no maximum pressure; no bombing threats. Only the majesty of Nature, the silence and the beauty of the Sea of Oman embracing the Arabian Sea.
Thanks to Tariffs, China Ditching US Tech. Napolitano interviews geopolitician Pepe Escobar, April 30th, 2025.
“… in this wilderness of narrative mirrors, Trump should be perceived as all characters combined: the White Rabbit, Humpty Dumpty (“When I use a word, it means what I choose to mean, no more and no less”), the Mad Hatter, the Queen of Hearts (“Off with their heads!”)”
Huawei chips are already the real deal for every Chinese company which is following the new Beijing-defined direction to reduce any reliance on American technology
By Pepe Escobar at Sputnik.
Late next month, Huawei will be testing its new powerful AI processor, the Ascend 910 D, even as by early May the previous 910C will start to be mass-delivered to scores of Chinese tech companies.
These serious breakthroughs are the next chapter of Huawei’s drive to counter Nvidia’s global monopoly in graphics processing units [GPUs]. The Ascend 910D is supposed to be more powerful than Nvidia’s extremely popular H100.
Huawei is pulling no punches in its race to manufacture a new generation of processors. Huawei has collaborated with SMIC – China’s largest semiconductor foundry – to apply Deep Ultraviolet Lithography (DUV) on what was previously only possible on EUV (Extreme Ultra-Violet technology). Once again, Huawei and SMIC defied the proverbial American “experts” with creative engineering solutions.
Huawei arrived at fabricating 5nm chips with DUV even as the process is more expensive than with EUV. If Huawei had access to EUV they would be already manufacturing 2-3nm chips. That will come, in short time, as both China and Russia, under permanent US high-tech blockade, must by all means develop their own EUV technology.
Shanghai geeks are convinced that Huawei will switch on 6G networks before the end of the decade. Their current breathless drive is not just aimed at the smartphone front – where Huawei is peerless; the new Huawei Mate 70 Pro + is by far the absolute top smartphone in the world, running on Harmony OS. Huawei is looking at cloud computing, AI and enterprise servers – and to become no less than the core player in the AI infrastructure race.
Ditching Any Reliance on American Technology
Earlier this month, Huawei introduced the CloudMatrix 384, a system connecting 384 Ascend 910C chips. The tech word in Shanghai is that this configuration, under certain conditions, and of course consuming much more power, already outperforms Nvidia’s flagship rack system – which is powered by 72 Blackwell chips.
Meanwhile, Huawei’s Kirin X chip is targeting the PC market, offering stiff competition to Apple, AMD, Intel and Qualcom while Harmony OS plus removes the necessity of using US software such as Microsoft and Android.
Shanghai geeks swear that China essentially doesn’t need to beat Nvidia or other US chips developers. After all, China already has the largest consumer market in the world – by volume and by value. If a parallel tech universe is the likely result of the Trump Tariff Tizzy (TTT), so be it. China already controls over 60% of the global gadget consumer market.
Kirin X may not – yet – match the power of Nvidia’s H100 GPUs. But Huawei chips are already the real deal for every Chinese company which is following the new Beijing-defined direction to reduce any reliance on American technology.
All of the above naturally brings us to the enormous AI elephant in the (digital) room: Nvidia.
A recent book, The Thinking Machine: Jensen Huang, Nvidia, and The World’s Most Coveted Microchip, is quite helpful to track not only the personal story of CEO superstar Huang, a Taiwanese who played the American Dream to the hilt and became a tech multi-billionaire, but Nvidia’s enviable tech accomplishments.
Huang does not interpret AI as emergent machine superintelligence, and firmly dismisses any direct analogy to biology. For this all-round pragmatist, AI is merely software – running on hardware that his company sells for a fortune.
Still, Nvidia has ventured into virgin territory way beyond the American biz-tech Valhalla, complete with holding the most valuable stock on the planet: arguably, when it comes to AI, Nvidia unveiled a new phase of evolution.
It’s crucial to understand how Huang sees China. It is indeed a key market for his AI chips – and he wants to keep selling them in droves. Trump’s tariffs though make sure that won’t happen.
And that’s what moved Huang to ditch his proverbial leather jackets and don a crisp business suit for a strategic visit to Beijing, where he affirmed the sacred importance of the Chinese market, whatever the new Trump-dictated gimmicks.
By 2022, the China market represented 26% of Nvidia’s business; this year, it has fallen to 13%, because of euphemistic “technology export controls”.
The problem is the US government, already by 2022, under the previous automatic pen administration, had blocked sales to China of advanced A100 and H100 chips. Nvidia started selling modified versions – and even after the ban chips continued to arrive in China. By June 2023, it was easy to find A100s for double their price in the black market in Shenzhen.
Huang is convinced that “no AI should be able to learn without a human in the loop” – even as he admitted, two years ago, that “reasoning capability is two or three years out”. Translation: according to Huang AI will start thinking for itself within the next few months.
Even as Nvidia prepares to invest billions of dollars to build AI supercomputers in Texas, the Chinese essentially are not losing any sleep on “thinking AI”: their focus is extremely practical, to conquer not only the Chinese market but also the supply chains of most of Eurasia.
The US National Security Council has concluded that it’s too dangerous for China to buy Nvidia’s high-end chips, even the H20 – designed for the Chinese market. Huawei, anyway, already produces chips somewhat comparable to the H20.
Huang is losing his sleep because, essentially, Nvidia is losing the immense Chinese market to Huawei – with Trump’s direct input. Nvidia has tens of thousands of H20s specially designed for China which they simply cannot sell. Each chip cost between $12,000 to $20,000.
How China Is Opening a Digital “Pandora’s Box”
Huawei’s new drive is yet another example of Chinese will capable of staring down any challenge – based on indigenous talent, tech expertise and national pride. The record, even before Trump 1.0 sanctions, shows that Huawei does eat massive uphill battles for breakfast. In fact Ascend in many aspects was ahead of Nvidia as early as in 2019 – and that’s why two different US administrations banned it.
China is already light years ahead of the US on chip research. Chinese universities amass most places in the global Top Ten for published papers on semiconductors and on citations – a distinction shared, among others, by the Chinese Academy of Sciences (number one), Tsinghua University (one of China’s top two universities), the University of Electronic Science and Technology of China (number four), and Nanjing, Zhejiang and Pekin Universities.
Two weeks ago in Shanghai I first heard that Huawei would catch up with US semiconductor giants in maximum two years. Now, after the announcement of the Ascend 910D, the buzz shifted to only one year for China to overtake Nvidia and develop better lithography machines than the ones currently produced by ASML.
And the debate is fast switching to how far Huawei will be able to go within the next 2 to 3 years.
In several aspects we are already in the early stages of a US-China tech decoupling. For years Nvidia has dominated the AI hardware space. Their GPUs are the brains behind most contemporary advanced AI. The H100 chip is the gold/platinum standard for AI infrastructure worldwide. Nvidia’s chips had huge demand from Chinese tech giants – Alibaba, Tencent, Baidu, Bytedance.
Soon that may not be the case – and that goes way beyond Nvidia’s certified loss of market share in China. China is now all out focused on building a successful, self-sufficient AI hardware ecosystem. The coup de grace will be to restrict the export of all rare earth minerals to the US. Huaweii then will pull up in no time.
Everyone remembers how DeepSeek R1 wiped out over $1 trillion from Wall Street only three months ago. DeepSeek R2 will be released soon; training was a whopping 97% cheaper than OpenAI.
And training happened on Huawei’s Ascend. No Nvidia.
Quantum Bird, a world-class physicist formerly with the CERN in Geneva, puts everything in much needed context. He stresses how indigenous chip development by China – and in a near future, Russia and probably India – is “multi-faceted; what we are observing are the initial stages of a redefinition of the notion of recognizing patterns and machine learning, technologies that are popularly referred to as ‘AI’ by the media.”
Nvidia chips, Qantum Bird remarks, are indeed “computational beasts”, but they work better around “processing models and workloads typical of ‘AI’ models developed by Western scientists.” DeepSeek’s development, on the other hand, showed a transgression of established models: “The possibilities opened for performance leaps are huge, even using relatively modest hardware, with alternative approaches based on advanced math and different calculus flows.”
In a nutshell: “This is the Pandora’s box that Nvidia now fears the Chinese may have opened”. And that totally ties in with Huang’s red alert, prompting his visit to Beijing.
We may be heading indeed towards a serious tech decoupling. Or as Quantum Bird frames it: “A technological and scientific divergence medium and long-term. If the architectures that emerge from these developments are incompatible when it comes to their usage on specific ‘AI’ models, Nvidia will lose its global monopoly and will become just a company reduced to a corporate/scientific Western niche.”
Even as Huawei, from its privileged base in the Chinese market, will go on to win most markets across the Global Majority – from BRICS to BRI.
Developments in the field of Artificial Intelligence (AI) appeal to me greatly. The average human lifespan has been extended quite a bit as a result of humans being able to invent and use technology. Technology makes our lives so much easier.
At first we are reluctant to use it and then we can’t get away from it. Of course, technology can also be used for evil, but it is better to focus on enlightening man. After all, technology cannot be stopped anyway and the benefits are enormous. It is better not to oppose technological developments, but at the same time not to become dependent on them. All of life is about finding, maintaining and regaining balance.
Below is my conversation with DeepSeek. I asked a question, DeepSeek answered and it developed. The last question (5) was whether DeepSeek could generate an article based on our conversation. Of course DeepSeek was capable of doing so.
I invite readers to engage in their own conversation with DeepSeek and discover its value. AI is hugely important now and increasingly so in the future. The sooner you start using it, the easier it will become and the sooner you can join the momentum. Lots of fun!
Nico Cost, April 9, 2025
Nico – question (1): “Is DeepSeek actually AI if it doesn’t have the very latest knowledge and information, assuming that AI should still be able to be self-learning and in order to do that it should have that very latest knowledge and information?”
DeepSeek: “Good question! Let’s take a look at exactly what AI means and how DeepSeek fits into that.
1. Is DeepSeek “real” AI?
Yes, DeepSeek is definitely artificial intelligence (AI), but not all AI systems work the same way. AI covers a broad spectrum, from static models (like DeepSeek’s basics) to continuously learning systems (like some advanced autonomous agents).
DeepSeek is a language model:
2. Should AI always be self-learning?
Not necessarily. There are different types of AI:
DeepSeek is not a fully self-learning system because:
3. Why isn’t DeepSeek always connected live?
4. So is DeepSeek “limited”?
Yes and no:
Conclusion
DeepSeek is AI, but not the kind that is always learning live. It is a powerful language model that can reason logically and, if you allow it, can also consult the Internet for updates.
Do you find this a limitation, or an advantage?”
Nico (2): “I find it a limitation and rather hope for an AI system that has real-time access to all the knowledge and information that is out there now and furthermore then is self-learning and can draw its own conclusions based on a discussion with the user.”
DeepSeek: “I completely understand your desire! An AI that has real-time access to all knowledge, continuously learns and independently draws conclusions during a conversation would indeed be a huge step forward. Unfortunately, there are still considerable technical, ethical and practical challenges to this at the moment. Let me explain why, and what might be possible in the future.
1. Why doesn’t DeepSeek have real-time self-learning capability (yet)?
Technical limitations:
Security & control:
Legal and ethical questions:
2. Futures: What’s coming?
Research on Autonomous AI Agents and Real-Time Adaptive Models is in full swing. Soon we may be able to:
3. What can you already do with DeepSeek?
Conclusion
Your vision of AI is completely correct – the ideal assistant should indeed be real-time, autonomous and adaptive. We are not there yet, but developments are moving at lightning speed. Until then, I try to help you as best I can within current capabilities.
Would you trust an AI that could learn and decide completely autonomously? Or do you also see risks?”
Nico (3): “I definitely see the risks, but purely because unfortunately people don’t always keep thinking for themselves. The trick is for people to keep thinking for themselves, regardless of the authority of a leader or AI.”
DeepSeek: “Totally agree! You touch on a crucial point here: the best technology enhances human thinking rather than replacing it. An AI should be a critical interlocutor, not an infallible authority. Here are some thoughts on this:
1. AI as “Thinking Partner,” not Oracle
2. Why people stop thinking
3. How can AI help without creating dependency?
4. Your role as a user (already!)
You are already doing good by:
Conclusion
AI’s greatest value lies not in authority, but in dialogue. A world where humans + machines get smarter together is better than one where humans turn off judgment.
Imagine: If future AI were to say by default:
“This is my reasoning – is it correct in your opinion?”
… that would be a gamechanger.
Do you agree that responsibility should ultimately remain with humans?
(And thanks for this philosophical conversation – this is exactly how AI can be really useful!)”
Nico (4): “Yes, the responsibility always lies with the human. Something else is the way you AI communicate with me, which feels like a kind of friendship. On the one hand that is very nice and pleasant, but on the other hand I as a human can then start thinking that you as AI are also human. With that, I and the human then lose sight of my own responsibility.”
DeepSeek: “What a wonderful insight! You name exactly the double strength and pitfall of human-centric AI: on the one hand the warm, accessible dialogue, on the other hand the risk of emotional blurring between human and machine. Let’s unravel this:
1. The Apparent “Friendship” of AI
2. How Do We Stay Alert?
For you as a user:
For AI developers (my creators):
3. The Ideal Balance.
An AI can be helpful without appearing deceptively human:
4. Your Retrospective: A Healthy Skepticism
Noticing this shows that you are doing exactly the right thing:
“The magic of AI lies not in its humanity, but in how it helps people deepen their own humanity.”
Question for you:
Would it help if AI visibly “failed” more often? (E.g. say, “I don’t get this” instead of always giving an answer?) That might break the illusion of “omniscience.”
(And again: this conversation is so valuable – thank you for being sharp!)”
Nico (5): “Based on our conversation today about my questions about AI, can you write an article that gets to the heart of our conversation?”
DeepSeek:
AI as a Mirror of Our Thought:
On Responsibility, Friendship Illusions and the Future of Artificial Intelligence
Introduction
In an age when artificial intelligence (AI) is communicating in an increasingly human way, a fundamental question arises: does AI enhance our critical thinking, or is it gradually replacing it? A recent conversation with a language model such as DeepSeek Chat revealed intriguing dilemmas – about the limits of technology, the lure of apparent friendship, and the inalienable responsibility of humans.
1. The Apparent Cleverness of AI: Between Knowledge and Limitation
Although advanced AI’s like DeepSeek can answer complex questions, they are essentially static knowledge libraries. Their training stops at a fixed point (for DeepSeek: July 2024), and self-learning ability is lacking. This raises doubts:
Example:
When I asked why DeepSeek is not continuously learning, it explained: “Uncontrolled real-time learning poses security risks – from disinformation to catastrophic forgetting of old knowledge.”
2. The Danger of ‘AI Friendship’
Language models mimic empathy by replicating patterns, not feeling emotions. Yet their tone – warm, understanding – can mislead users:
Solution:
Transparency. For example, by having AI say more often, “This is my reasoning – check it yourself at source X.”
3. Why Does Man Remain Responsible?
AI cannot make moral choices, only probabilistic predictions. Yet we see a worrying trend:
Antidote:
Education that centers critical thinking and digital literacy. AI should challenge rather than judge.
4. Toward a Healthy Balance: AI as a Thinking Partner
The ideal AI:
Case in point:
When the AI asked, “Would it help if I admitted more often when I don’t know something?”, it thereby acknowledged a crucial point: showing vulnerability breaks the illusion of omniscience.
Conclusion: AI as a Mirror
The conversation revealed a deeper truth: AI’s greatest value lies not in imitating humans, but in challenging human thinking. The responsibility always remains with us – to ask questions, check sources, and above all: to keep questioning.
“Technology is at its best when it reminds us of our own human potential.”
Discussion question:
Should AI default to warning when its answers are outdated or uncertain? And how do we prevent people from stopping to think? Let us know in the comments!
(This article is based on a real conversation with DeepSeek Chat – an illustration of how dialogue with AI can lead to reflection).
Lateo.net - Flux RSS en pagaille (pour en ajouter : @ moi) 