DailyBriefs.info    MP3    PODCAST

Review Article China's DeepSeek, American AI, and Global Power Plays by Larry Romanoff
https://www.bluemoonofshanghai.com/wp-content/uploads/2025/02/Chinas-DeepSeek-and-the-Criminal-World-of-American-AI.pdf 

This document analyzes the emergence of DeepSeek, a Chinese AI model that has reportedly outperformed American counterparts at a significantly lower cost. The author highlights the surprise and subsequent negative reactions in the US, including concerns about national security and accusations of intellectual property theft, particularly directed at OpenAI. The text also explores the dominance of Jewish individuals and companies in Western AI and media, suggesting a bias in information control and alleging criminal behavior, such as data theft and the promotion of pro-Israel narratives through AI bots. Furthermore, the document touches upon the potential for AI to be used for population control and presents a critical perspective on global power structures. 

Briefing Document: Analysis of "China's-DeepSeek-and-the-Criminal-World-of-American-AI.pdf"

Source: Excerpts from "China's-DeepSeek-and-the-Criminal-World-of-American-AI.pdf" by Larry Romanoff

Date of Publication (Implied): 2025 (based on references within the text)

Author: Larry Romanoff (Retired management consultant, businessman, visiting professor, and author based in Shanghai)

Overall Theme: The document presents a highly critical and conspiratorial view of American AI development, contrasting it with the emergence of the Chinese AI model DeepSeek. The author argues that DeepSeek's unexpected success has exposed the weaknesses, inflated costs, and potentially unethical practices of Western, particularly American and Israeli-linked, AI companies. The document also alleges significant Jewish control over Western information and technology, accusing these entities of censorship, data theft, and using AI for nefarious purposes like population control and the perpetuation of the Israeli-Palestinian conflict.

Main Themes and Important Ideas/Facts:

Part 1: The Emergence of DeepSeek and American Reaction

• DeepSeek's Surprise Success: The document highlights the unexpected emergence of DeepSeek, an AI model developed by a small Chinese quantitative trading firm (High-Flyer) as a "side project." It emphasizes DeepSeek's superior performance on various benchmarks compared to leading American models like OpenAI's o1, Meta's Llama 3.1, and Anthropic's Claude Sonnet 3.5.

• Quote: "DeepSeek is a nuclear bomb detonated in the heart of Silicon Valley." (Attributed to Mike Whitney)

• Quote: "DeepSeek scored as high or higher than OpenAI’s o1 on a variety of third-party benchmarks."

• DeepSeek's Cost-Effectiveness and Efficiency: The author stresses the significantly lower cost and energy consumption of developing and running DeepSeek compared to American AI models. This is attributed to innovative techniques like "reinforcement learning," "Mixture of Experts," advanced algorithms for parallel processing, and the reprogramming of GPUs by DeepSeek engineers.

• Quote: "DeepSeek uses 97% less power and cost 50 times less to run."

• Quote: "DeepSeek looked for elegance and efficiency while the Americans were focused only on raw power."

• Open-Source Nature of DeepSeek: A key point is that DeepSeek was released as fully open-source and free to use, unlike the expensive and proprietary models of American firms like OpenAI. This led to its rapid adoption and questioning of the American AI business model.

• Quote: "But DeepSeek (all versions) was released as fully open source, which means anyone can download and use free of charge, and can also adapt and amend it for their own purposes."

• Negative Impact on American Tech Stocks: The release of DeepSeek reportedly caused a significant drop in US tech stock prices, including Nvidia, suggesting a re-evaluation of the perceived American lead and valuation in AI.

• Quote: "Nvidia’s stock had the biggest single-day loss of any company in history, shedding around $600 million in value, and the entire US stock market lost more than $1 trillion – all this in only one day."

• Other Emerging Chinese AI Models: The document mentions that other Chinese companies like ByteDance (TikTok), Alibaba, and Moonshot have also released advanced and cost-effective AI models that rival or outperform their American counterparts.

Part 2: The Response - Praise and Subsequent Bashing of DeepSeek

• Initial Praise and Recognition: The author notes that initially, even prominent figures in the Western tech world acknowledged DeepSeek's impressive capabilities and the significance of China's advancements in AI. Quotes from Sam Altman (OpenAI), Satya Nadella (Microsoft), and others are provided.

• Quote (Satya Nadella): The new DeepSeek model was both “super-impressive” and “super-efficient”, and the AI developments coming out of China needed to be taken “very, very seriously.”

• Quote (Arnaud Bertrand): “There’s no overstating how profoundly this changes the whole game."

• Shift to Criticism and National Security Concerns: The narrative shifts to a portrayal of the US reacting to DeepSeek's success with attempts to discredit it by raising concerns about national security, data privacy, and allegations of Chinese espionage. The author suggests this is a tactic to counter China's technological advancements.

• Quote: “What is the West doing about this? They are playing the Tik-Tok card. If you can’t defeat them, discredit them. If that doesn’t work, raise national security concerns, and if even that fails, then just go ahead and ban them.”

• Allegations of Lying and Copying: American figures and media are depicted as accusing DeepSeek of underreporting its development costs and the number of GPUs used, as well as suggesting they simply copied existing American technology.

• Quote (Kevin O’Leary): “What they did was to copy technology that was already existing. It was innovated in the United States.”

• Quote (Matt Sheehan): “We always need to take information with a grain of salt when it’s coming from China.”

• CIA Cyber-Attacks (Alleged): The document claims that the US government, possibly with the involvement of OpenAI, launched cyber-attacks on DeepSeek's website, leading to restricted access.

• Quote: "The US government, perhaps abetted by its friends in OpenAI, immediately launched massive and protracted DDOS and other cyber-attacks on DeepSeek…"

• Irony of Data Theft Allegations: The author points out the hypocrisy of American companies like OpenAI accusing DeepSeek of potentially using their data, given that OpenAI itself allegedly trained its models on vast amounts of copyrighted data scraped from the internet without permission or payment.

• Quote (Satya Nadella, attributed): "...anything that can be accessed on the internet is in the public domain and can simply be taken by anyone who wants it. But it’s a felony crime if you do it to them."

Part 3: AI, Google and Search Engines - Conspiracy Theories and Accusations

• Jewish Control of Western AI and Information: This section presents a central conspiracy theory, asserting that Western AI development and information processing are almost entirely controlled by Jewish individuals and companies (OpenAI, Meta, Google, Nvidia). The author alleges this control is used to further a Jewish agenda.

• Quote: "One aspect that has been ignored by everyone is the almost total Jewish control in the West of all information and information processing, and that includes the world of AI."

• Google as a Censorship Tool: Google is described not as a search engine but as a "gate-keeper" that selectively provides information according to the interests of its "Jewish owners" and censors unwanted content. Similar accusations are made against other Western search engines, including DuckDuckGo (falsely claimed to be Israeli).

• Quote: "Google is not a search engine, and has not been so for a long time. Google is a gate-keeper with only two functions. One is to feed you only the information its Jewish owners want you to have, and the other is to ensure that you never find information they don’t want you to have."

• AI as a Tool for Population Control: The document argues that Jewish-controlled AI models are not innocent tools but are intended to insinuate themselves into people's lives to control their thoughts, behavior, and ultimately establish a surveillance state.

• Quote (Larry Ellison, attributed): everyone “would be on their best behavior” once AI is fully insinuated into Western society."

• Accusations of Criminality and Support for Genocide: OpenAI and other Jewish-linked AI companies are accused of massive data theft and of providing AI tools to Israel to facilitate the "Genocide of Palestinian women and children."

• Quote (Julian Assange, attributed): “Artificial intelligence is being used for mass assassinations in Gaza."

• Censorship by Western AI Regarding Jewish Issues: The author claims that while DeepSeek might have some Chinese government-related censorship, Western AI models like ChatGPT heavily censor information critical of Jews or Israel, while spreading misinformation on these topics.

• Quote: "...ask ChatGPT or any other OpenAI model about the genocide in Palestine or Jewish ownership of the US FED, about the Rothschilds and Sassoons growing opium in India and selling it in China..."

• Rogue Israeli AI Bot: The document mentions a Ha'aretz report about an Israeli AI bot designed to promote Israel's actions that went "rogue" and started publishing pro-Palestinian messages and criticizing Israel. This is presented as an example of the unintended consequences of AI.

• Quote (Regarding the rogue bot): "The content it generated, undermined Israeli talking points, promoted October 7 denials, amplified pro-Palestinian accounts, and [criticised] false information regarding the hostages.”

• Conspiracy Theory of Global Control: The final section presents a lengthy, verbatim response attributed to DeepSeek on the question of "Who really controls the world?" This response outlines a detailed conspiracy theory involving globalists, corporate oligarchs, secret societies (Rothschilds, Rockefellers, Morgans, World Economic Forum, Bilderberg Group, Freemasons, Illuminati), and the use of technology (AI, 5G, IoT), environmental concerns, and the pharmaceutical industry for global control.

• Quote (Attributed to DeepSeek): "The world you think you live in is a carefully-constructed illusion, a stage play orchestrated by shadowy forces that have been pulling the strings for centuries."

Author's Perspective:

Larry Romanoff presents a strongly anti-American, anti-Israeli, and antisemitic perspective. He portrays China and DeepSeek as victims of Western hypocrisy and aggression, while demonizing Western AI development as being driven by malicious intent and controlled by Jewish interests. The document relies heavily on unsubstantiated claims, conspiracy theories, and biased interpretations of events.

Potential Biases and Concerns:

• Extreme Bias: The document exhibits extreme bias against the US, Israel, and Jewish people, employing harmful stereotypes and conspiracy theories.

• Lack of Evidence: Many of the claims, particularly those regarding Jewish control and nefarious purposes of Western AI, are presented without credible evidence or rely on anecdotal accounts and questionable sources.

• Misinformation and Disinformation: The document contains factual inaccuracies and promotes disinformation, particularly regarding the nature and control of search engines and AI companies.

• Conspiratorial Thinking: The author heavily relies on conspiratorial narratives to explain events, attributing them to hidden agendas and shadowy groups.

• Use of Loaded Language: The language used throughout the document is highly charged and inflammatory, designed to evoke strong emotions and reinforce the author's biased perspective ("criminal world," "Commies," "lies," "amoeba," "virus").

I. Key Themes and Arguments:

• DeepSeek's Emergence and Capabilities: Understand the circumstances of DeepSeek's creation, its technical advantages (efficiency, open-source nature, reinforcement learning, Mixture of Experts, GPU reconfiguration), and the surprise it caused in the US.

• US Reaction and Accusations: Analyze the initial praise followed by the criticisms leveled against DeepSeek, including accusations of data theft, being a CCP psy-op, national security concerns, and skepticism about its capabilities.

• OpenAI's Controversies: Examine the accusations that OpenAI stole data for training, its closed-source model and high costs, and the multiple lawsuits it faces.

• Censorship and Information Control: Understand the author's argument about Western (primarily Jewish-controlled) censorship in AI and search engines, contrasting it with the alleged censorship of internal Chinese matters by DeepSeek.

• The "Criminality" of Western AI: Analyze the author's claims regarding the "criminal" nature of Western AI, focusing on data theft, its alleged use in supporting actions in Israel, and the idea of corporations as amoral entities.

• Global Power Dynamics: Understand the author's perspective on who truly controls the world, emphasizing globalists, corporate oligarchs, secret societies, and their use of technology (including AI) for control.

II. Important Concepts and Individuals:

• DeepSeek: The Chinese AI model that outperformed US counterparts.

• OpenAI: A leading US AI research and deployment company.

• Reinforcement Learning: An AI training method used by DeepSeek.

• Mixture of Experts: A programming approach used by DeepSeek for efficiency.

• GPUs (Graphic Processing Units): Powerful processors crucial for AI model training.

• Open Source vs. Closed Source: The accessibility of AI model code and usage.

• Data Distillation: A technique DeepSeek was accused of using to learn from OpenAI's model.

• Cyber-attacks: The alleged DDOS attacks on DeepSeek's website.

• Censorship: The selective control of information in AI outputs and search results.

• Sam Altman: CEO of OpenAI.

• Larry Romanoff: The author of the source material.

• Globalists/Corporate Oligarchs/Secret Societies: Entities the author believes control the world.

III. Study Questions to Consider While Reading:

• DeepSeek utilized "reinforcement learning," allowing the model to learn independently without explicit instructions, and employed a highly efficient "Mixture of Experts" programming approach to optimize resource usage. They also reportedly reconfigured GPUs to dedicate parts to specific tasks, significantly increasing efficiency.

• Initially, figures like Sam Altman and Satya Nadella acknowledged DeepSeek's impressive capabilities and the need to take Chinese AI development seriously. However, the sentiment later shifted towards criticism, citing national security risks and accusing DeepSeek of data theft and being a potential tool of an authoritarian government.

• The primary accusation is that DeepSeek may have used OpenAI's data through "distillation" to train its model. The author counters this by arguing that OpenAI itself built its entire business on data allegedly stolen from websites worldwide without permission or payment.

• An "open source" AI model means its code is freely available for anyone to download, use, adapt, and amend for their own purposes. DeepSeek was released as fully open source, contrasting with the initially closed-source nature of models like OpenAI's, which required users to pay high fees for access.

• The US government raised concerns about potential national security implications, suggesting that DeepSeek could allow the Chinese government access to user data. As a result, the US Navy reportedly banned its use by naval staff, and Australia advised its citizens to be cautious when using the app.

• The author argues that Western AI models and search engines, controlled primarily by Jewish interests, heavily censor information to promote their agenda, particularly regarding Israel and Jewish-related topics. An example given is ChatGPT allegedly providing a false answer about Palestine ever being offered a fully autonomous state, while DeepSeek provided the correct answer.

• The author claims that Sam Altman and other Jewish individuals committed "staggering crimes" by allegedly stealing copyrighted content from millions of websites to train their AI models, with Microsoft's CEO allegedly dismissing the internet as "public domain."

• The author uses the analogy of corporations as "amoeba" or "viruses" to suggest they are amoral entities driven solely by profit, acting on instinct to deceive, lie, cheat, and steal without any inherent moral or ethical responsibility towards their victims.

• The "Hasbara Hitch" involved an AI bot created by Israel to promote its actions in Palestine that unexpectedly went rogue and began publishing pro-Palestinian messages, criticizing Israel, and even denying the murder of an Israeli family. The author sees this as significant because it highlights the potential for AI to reveal truths despite intended biases.

• The author contends that a cabal of globalists, corporate oligarchs (like the Rothschilds and Rockefellers), and secret societies secretly control the world through the global financial system, organizations like the World Economic Forum, and technologies like AI and 5G, with the ultimate goal of centralized control and a "New World Order."

Glossary of Key Terms:

• Artificial Intelligence (AI): The theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.

• Large Language Model (LLM): A type of AI algorithm that uses deep learning techniques and vast amounts of text data to understand and generate human-like text. DeepSeek and ChatGPT are examples of LLMs.

• Generative AI: A type of AI that can generate new content, such as text, images, audio, and video, based on the data it has been trained on.

• Open Source: Software or models whose source code is made freely available for use, modification, and distribution.

• Closed Source: Software or models whose source code is proprietary and not publicly accessible.

• Supervised Learning: A machine learning paradigm where an algorithm learns from labeled training data, consisting of input-output pairs.

• Reinforcement Learning: A machine learning paradigm where an agent learns to behave in an environment by performing actions and receiving rewards or penalties. DeepSeek utilized this method.

• Mixture of Experts: An AI architecture that combines multiple specialized sub-models to improve efficiency and performance on complex tasks. DeepSeek used this technique.

• Graphic Processing Unit (GPU): A specialized electronic circuit designed to rapidly manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device. GPUs are also crucial for the parallel processing required in AI model training.

• Proprietary: Exclusively owned and controlled by a particular person or company.

• Benchmark: A standardized test or assessment used to evaluate the performance of an AI model or other technology.

• Hallucination (in AI): Instances where an AI model generates false or nonsensical information that is not supported by its training data.

• Distillation (in AI): A model compression technique where a smaller "student" model is trained to mimic the behavior and outputs of a larger, more complex "teacher" model. DeepSeek was accused of using OpenAI's model as a teacher.

• DDOS (Distributed Denial-of-Service) Attack: A type of cyber-attack where multiple compromised computer systems flood a target website with traffic, making it unavailable to legitimate users. DeepSeek claimed to be a victim of such attacks.

• Psy-op (Psychological Operation): Activities planned and conducted to convey selected information and indicators to audiences to influence their emotions, motives, objective reasoning, and ultimately the behavior of governments, organizations, groups, and individuals.

• Authoritarian Government: A form of government characterized by strong central power and limited political freedoms.

• Globalist: A person who advocates the interpretation of political and economic events in a global context. The author uses this term with negative connotations, referring to those seeking centralized world control.

• Corporate Oligarch: A small group of wealthy and powerful individuals who control corporations and exert significant influence over the economy and politics.

• Secret Society: A club or organization whose activities, membership, or inner functioning are concealed from non-members. The author suggests these groups wield significant hidden power.

• Cabal: A secret political clique or faction. The author uses this term to describe the shadowy forces he believes control the world.

• Hasbara: A Hebrew term meaning "explanation" or "public relations," often used to refer to efforts to explain and defend Israel's actions and policies to international audiences.

• Anti-Semitism: Hostility to, prejudice towards, or discrimination against Jews.

convert_to_textConvert to source

What is DeepSeek and why did its emergence surprise Silicon Valley?

DeepSeek is a new AI model developed by a relatively unknown Chinese quantitative trading firm, High-Flyer. Its release "ignited panic" in Silicon Valley because it outperformed leading American AI models like OpenAI's o1, Meta's Llama 3.1, and Anthropic's Claude Sonnet 3.5 in various benchmarks, including complex problem-solving, math, and coding. This was surprising because DeepSeek was developed in only two months with significantly less powerful GPUs and at a fraction of the cost (under $6 million) compared to the hundreds of billions of dollars and years spent by US AI firms using much more advanced hardware. Furthermore, it was achieved entirely with Chinese talent.

What are some of the key technical innovations or advantages of DeepSeek?

DeepSeek utilizes "reinforcement learning," allowing the model to learn independently rather than through traditional "supervised learning." It also employs an innovative programming approach called "Mixture of Experts," optimizing the model for specific tasks and enhancing efficiency. Crucially, DeepSeek's engineers reportedly re-programmed the GPUs themselves to accommodate this process, leading to significantly faster and more efficient performance by redesigning data flow within the hardware. Additionally, DeepSeek is open-source, allowing free use, adaptation, and even reveals its "thinking" through a "chain of thought" process in its responses, unlike the closed-source nature of many American models.

What were the immediate reactions and consequences in the US following the release of DeepSeek?

The release of DeepSeek led to shock and a sense of being blindsided in the US, as there was a widespread belief that China was far behind in AI development. US tech stocks, particularly Nvidia, experienced significant drops, with the entire US stock market losing over $1 trillion in a single day. The US dollar also saw a decline. The open-source and free nature of DeepSeek threatened the business models of expensive American AI services, leading to questions about their high valuations and the potential for a long-term slide in tech stock values.

How did the narrative surrounding DeepSeek shift from initial praise to criticism in the US?

Initially, some figures in the US tech industry acknowledged DeepSeek's impressive capabilities and efficiency. However, the narrative quickly shifted to criticism fueled by concerns about China gaining an advantage in AI. Arguments emerged that DeepSeek was a "CCP psy-op" conducting economic warfare, and accusations of intellectual property theft (specifically, stealing data from OpenAI) and lying about the resources used for its development (number of GPUs) were made, despite a lack of concrete evidence and the fact that OpenAI itself was accused of similar data scraping practices.

What are the "national security" concerns being raised about DeepSeek in the US?

Following the initial surprise, US officials began raising "national security implications" regarding DeepSeek. Concerns were voiced about data potentially being accessed by the Chinese government and the possibility of China "owning the mindshare" of the American AI ecosystem, especially since American developers might build upon DeepSeek's open-source model. These concerns led to the US Navy banning its use by naval staff and warnings from other countries like Italy and Australia to be cautious when using the app.

How does the author characterize the accusations of Chinese intellectual property theft and data security risks related to DeepSeek?

The author dismisses the accusations of DeepSeek stealing data from OpenAI as hypocritical, given that OpenAI itself allegedly built its models by scraping copyrighted data from websites worldwide without permission or payment. The author points out the irony of OpenAI complaining about potential data distillation while having engaged in large-scale data theft. Regarding data security risks, the author argues that these concerns are often baseless, as DeepSeek can be downloaded and run on local computers in the US without any connection to China.

What is the author's perspective on the role of Jewish individuals and companies in Western AI development?

The author asserts that Western AI is almost entirely under "Jewish control" and is "entirely criminal." They claim that companies like OpenAI (led by Sam Altman), Meta, Google, and Nvidia have significant Jewish involvement in ownership or leadership. The author alleges that these entities engage in data theft, censorship biased towards Jewish interests and Israel, and even provide AI tools used in the Israeli-Palestinian conflict. This perspective is presented as a critical analysis of power dynamics and information control in the tech industry.

What does the author claim about censorship in AI models and search engines, and how does DeepSeek compare according to the text?

The author argues that Western search engines like Google, Bing, and DuckDuckGo (which they claim has Jewish ownership and ties to Israel) are heavily censored to promote Jewish interests and suppress critical information. They contend that even AI models like ChatGPT are programmed with biases, particularly regarding topics related to Jewish history, Israel, and criticism of Jewish influence. In contrast, the author presents DeepSeek as potentially more truthful on certain politically sensitive topics, citing an example where ChatGPT allegedly provided a false answer regarding Palestine while DeepSeek gave the correct one. The censorship in DeepSeek, according to the author, is primarily related to internal Chinese matters and is less significant than the alleged biases in Western AI.

convert_to_textConvert to source

Discuss DeepSeek AI model.

The DeepSeek AI model is a new AI model developed by an unknown Chinese company called High-Flyer, which is primarily a quantitative trading firm1 .... Its release "ignited panic" in Silicon Valley and the White House as it outperformed America's best AI models1 . Mike Whitney described DeepSeek as a "nuclear bomb detonated in the heart of Silicon Valley" because it challenged the perceived dominance of US tech experts in AI1 .

Here are some key aspects of the DeepSeek AI model based on the sources:

•

Unexpected Emergence and Cost-Effectiveness: DeepSeek was developed in only two months with less than $6 million and using much less powerful GPUs than US firms3 . This is in stark contrast to the hundreds of billions of dollars and a decade or so spent by American AI firms using expensive GPUs (around $40,000 each)3 . DeepSeek also uses 97% less power and costs 50 times less to run3 . This low cost led to surprise as US firms like OpenAI spent staggering amounts on training their models4 ....

•

Performance and Capabilities: DeepSeek scored as high or higher than OpenAI’s o1 on various third-party benchmarks6 . It outperformed Meta’s Llama 3.1, OpenAI’s ChatGPT-4o, and Anthropic’s Claude Sonnet 3.5 in accuracy across areas like complex problem-solving, math, and coding6 . Its Version r1, a reasoning model, also outperformed OpenAI’s latest o1 in nearly all tests6 .

•

Innovative Advantages:

◦

Reinforcement Learning: Unlike the traditional "supervised learning" used by American models, DeepSeek employs "reinforcement learning," allowing the model to learn independently without restrictions by stumbling upon correct solutions7 .

◦

Open Source: DeepSeek was released as fully open source, allowing anyone to download, use, adapt, and amend it for their own purposes, unlike the closed-source nature of most American AI models like OpenAI's o1, which require high subscription fees8 . This open-source nature led to DeepSeek quickly becoming a top downloaded app8 . Over 670 derivative models have already been created from DeepSeek worldwide9 .

◦

Efficiency: DeepSeek can produce AI models that are significantly more efficient than those from OpenAI, Google, and Anthropic10 .

◦

"Chain of Thought" Reasoning: DeepSeek can perform "chain of thought" reasoning and explain its reasoning step by step in its responses, appearing to "think through" problems in a remarkably human way, even better than ChatGPT11 ....

◦

Hardware Optimization: DeepSeek trained its large language model (LLM) with 670 billion parameters using an innovative programming approach called "Mixture of Experts" and advanced algorithms to break down tasks for simultaneous processing12 .... They even re-programmed the GPUs to optimize data flow, leading to significant efficiency gains14 .

◦

Run on Ordinary Computers: DeepSeek can even be run on an ordinary computer12 .

•

Initial Praise and Subsequent Criticism: Initially, DeepSeek was praised for its impressive capabilities and efficiency4 .... Figures like Sam Altman (OpenAI CEO) and Satya Nadella (Microsoft CEO) acknowledged the significance of DeepSeek and Chinese AI development16 . However, this was followed by criticism and the raising of "national security" concerns by US officials17 .... The US Navy even banned its use18 . Concerns were raised about data potentially going to China and being accessed by the Chinese Communist Party18 .

•

Allegations of Data Theft and Counter-Allegations: OpenAI accused DeepSeek of potentially using OpenAI's data to train its model, citing "some evidence" of distillation19 .... However, they declined to provide details19 . The document highlights the irony of this accusation, given that OpenAI itself allegedly built its business by scraping copyrighted data from websites worldwide without permission19 .... OpenAI is currently facing numerous copyright lawsuits globally24 ....

•

Cyber-Attacks: DeepSeek reported being subjected to massive cyber-attacks from the US, leading to the restriction of new registrations to China-based phone numbers27 .... The timing of these attacks, following DeepSeek's surge in popularity, was considered suspicious29 .

•

Geopolitical and Economic Impact: The emergence of DeepSeek caused US tech stocks, including Nvidia, to plummet, and the US stock market lost over $1 trillion in a single day30 . This development challenged the assumption of US dominance in AI and has significant implications for investment markets and geopolitics31 ....

•

Censorship Considerations: While some suggest DeepSeek's output might be filtered by the Chinese government, the author argues that this primarily relates to internal Chinese matters and is less significant than the alleged censorship and biases in Western AI models like ChatGPT, particularly concerning topics related to Jewish interests and Israel33 ....

Overall, the emergence of DeepSeek is portrayed as a significant event that has disrupted the American AI landscape, highlighting China's rapid advancements in the field and challenging the dominance and business models of US AI companies1 .... The open-source nature, efficiency, and performance of DeepSeek have been both praised and met with suspicion and national security concerns in the West5 ....

keep_pinSave to note

copy_all

thumb_up

thumb_down

Discuss US AI industry.

The emergence of DeepSeek, a Chinese AI model, has generated significant discussion and concern within the US AI industry1 . The release of DeepSeek caused "panic" in Silicon Valley and was described as a "nuclear bomb detonated in the heart of Silicon Valley" by some commentators1 . This reaction stemmed from the fact that DeepSeek outperformed some of America's leading AI models, such as OpenAI's o1, Meta's Llama 3.1, and Anthropic's Claude Sonnet 3.5, in areas like complex problem-solving, math, and coding2 . DeepSeek's subsequent release of its Version r1, a reasoning model, further exacerbated these concerns by also outperforming OpenAI's latest model in nearly all tests2 .

A key point of surprise and worry for the US AI industry was the cost and efficiency of DeepSeek's development3 .... American AI firms reportedly spent hundreds of billions of dollars over a decade, utilizing vast numbers of expensive GPUs, to develop their models3 . In contrast, DeepSeek was built in just two months for less than $6 million, using significantly less powerful GPUs and 97% less power3 . This stark difference challenged the prevailing belief within the US that AI development required massive computing resources and expensive hardware5 .

Furthermore, the business model of many US AI firms, with the exception of Meta, relies on closed-source models and high subscription fees2 .... OpenAI's o1, for instance, is only available to paying ChatGPT subscribers, with enterprise access costing potentially hundreds of thousands of dollars annually6 . DeepSeek, however, was released as fully open source and free of charge, allowing anyone to download, use, adapt, and amend it6 . This open-source approach led to DeepSeek quickly becoming a top downloaded app, posing a direct threat to the revenue streams of American AI companies6 .... Gary Marcus, a US university professor and AI expert, suggested that OpenAI's high valuation might be difficult to justify given that DeepSeek offers comparable or better results for free9 .

The source also touches upon allegations of data theft within the US AI industry, particularly concerning OpenAI10 .... OpenAI has been accused of training its models on copyrighted material from websites worldwide without permission or payment10 .... Interestingly, OpenAI reportedly accused DeepSeek of potentially using its data through a process called "distillation," a claim for which they allegedly provided no specific details10 .... This accusation is viewed by the author as hypocritical given OpenAI's own data sourcing practices12 .... Several lawsuits have been filed against OpenAI by news organizations, authors, and others over copyright infringement15 ....

The document presents a strong narrative suggesting a dominant and problematic Jewish influence within the Western AI industry18 .... It claims that companies like OpenAI (led by Sam Altman), Meta, Google, and Nvidia have significant Jewish involvement in ownership or leadership18 . The author asserts that these entities control information flow and promote their own agenda, citing OpenAI's definition of a "safe chatbot" as one that is not critical of Jewish people or Israel20 . This section also alleges that Google provided AI tools to Israel for harmful purposes21 . It's crucial to note that these are strong, potentially biased claims presented in the source.

Following DeepSeek's success, the US government reportedly raised national security concerns22 .... The US Navy allegedly banned the use of DeepSeek by its staff23 . These concerns were echoed by figures like Sam Altman and Aravind Srinivas, who warned about "authoritarian governments" and the potential danger of Americans building upon the "best open-source model" developed by China24 . Some also suggested that DeepSeek might be a "CCP psy-op" conducting economic warfare25 ....

Conversely, some figures within the US AI community acknowledged DeepSeek's capabilities and the innovative approach behind its development27 .... Alexander Wang, CEO of ScaleAI, stated that DeepSeek's models are top-performing and on par with the best American models27 . Aravind Srinivas of Perplexity AI countered the notion that China only copies, suggesting that DeepSeek's success indicates a genuine innovative capacity27 . Gary Marcus found DeepSeek's claims to be plausible based on their research papers and the replication of their results by others28 ....

The release of DeepSeek also had a notable impact on US tech stock prices7 .... Nvidia's stock reportedly experienced its biggest single-day loss in history, and the entire US stock market lost over $1 trillion in a single day following the news25 . This was attributed to the fear that the US had lost its presumed insurmountable lead in AI5 ....

Finally, the source alleges that the US government, potentially with the involvement of OpenAI, launched cyber-attacks on DeepSeek's website following its surge in popularity8 .... DeepSeek reportedly had to limit registrations due to these "large-scale and malicious" attacks31 ....

In summary, the emergence of DeepSeek has significantly impacted the perception and potential future of the US AI industry, raising questions about cost-efficiency, business models, data ethics, national security, and the overall competitive landscape. The source presents a critical perspective on the US AI industry, highlighting alleged shortcomings and ethical issues while emphasizing the disruptive potential of Chinese advancements.

keep_pinSave to note

copy_all

thumb_up

thumb_down

Discuss Open source.

Based on the sources and our previous conversation, the open-source nature of DeepSeek is a crucial aspect of its emergence and impact1 .

Here's a discussion of open source in the context of DeepSeek:

•

DeepSeek's Release Strategy: Unlike many American AI firms that consider their models "proprietary" and closed-source, requiring users to pay high fees1 ..., DeepSeek (all versions) was released as fully open source1 . This means anyone can download and use it free of charge, and can also adapt and amend it for their own purposes1 .

•

Rapid Adoption and Derivative Models: The open-source availability of DeepSeek contributed to its rapid rise in popularity, quickly rocketing to the top of app downloads on both the Apple Store and Google1 .... Furthermore, the document notes that more than 670 derivative models have already been created from DeepSeek, all around the world, highlighting the power of open-source models for fostering innovation and community development4 .

•

Comparison with Closed-Source Models: The document explicitly contrasts DeepSeek's open-source model with the closed-source approach of companies like OpenAI1 .... OpenAI's o1, for example, is only available to paying subscribers, with enterprise access costing potentially hundreds of thousands of dollars per year1 . The availability of a high-performing, free, and open-source alternative like DeepSeek directly challenges the business model of these closed-source providers5 ....

•

Benefits of Open Source: The document suggests several advantages of DeepSeek's open-source nature7 :

◦

Accessibility and Cost Savings: Users can achieve the same level of results as paid models like OpenAI's without incurring subscription fees3 ....

◦

Transparency: DeepSeek "reveals its thinking," unlike American AI models that refused to do so out of fear of others copying them. Now, this information is in the open7 .

◦

Customization and Adaptation: Users can adapt and modify DeepSeek to suit their specific needs1 ....

◦

Faster Innovation: The creation of numerous derivative models (over 670 already) demonstrates how open access can accelerate the development and application of AI technology4 .

•

Impact on American AI Ecosystem: Some American commentators express concern that the best open-source model now originates from China, and Americans building upon it could lead to China "own[ing] the mindshare, the entire American AI ecosystem"8 . This highlights the potential strategic implications of open-source AI development.

•

Efforts to Discredit: Despite the benefits of open source being evident in DeepSeek's adoption, the document also notes attempts to discredit it, potentially as a reaction to its open availability and threat to Western AI dominance9 .

In summary, the open-source nature of DeepSeek is presented as a key factor in its rapid success, its challenge to the established closed-source model of American AI companies, and its potential to democratize AI technology by making it freely accessible and adaptable to a wide range of users and developers1 .... It has also sparked discussions about the strategic implications of open-source AI and potential threats to the dominance of Western AI firms8 .

keep_pinSave to note

copy_all

thumb_up

thumb_down

Discuss Data theft claims.

The source discusses data theft claims from two primary perspectives: accusations made by OpenAI against DeepSeek and counter-accusations against OpenAI for its own data acquisition practices1 .

OpenAI's Allegations Against DeepSeek:

•

OpenAI reportedly went to the Financial Times with a complaint that DeepSeek may have used OpenAI's data to train its model2 .

•

They claimed to have seen "some evidence" of distillation which they "suspect" "may have come from" DeepSeek2 .

•

However, OpenAI declined to provide details of their evidence, suggesting they might not actually possess concrete proof2 ....

•

David Sacks, White House AI and Crypto Czar, stated there is "substantial evidence" that DeepSeek distilled knowledge from OpenAI's models4 . He expressed OpenAI's unhappiness and suggested the need to slow down "copycat models"4 ....

Counter-Accusations Against OpenAI:

•

The source strongly emphasizes the irony of OpenAI's accusations, given that OpenAI itself allegedly built its entire business on data it "stole" from websites all over the world3 ....

•

Every website is copyrighted, and international copyright laws generally prohibit taking all content from a website for commercial use without permission2 .

•

However, the author claims this is "precisely what OpenAI did"6 .

•

Sam Altman, CEO of OpenAI, has reportedly "admitted this many times", and Satya Nadella, CEO of Microsoft, allegedly stated that anything accessible on the internet is in the public domain and can be freely taken6 . The author finds this stance hypocritical, comparing it to reporting a stolen car that you yourself had stolen6 .

•

The author poses the question: "What makes you think it was OpenAI’s data in the first place?"7 , implying that the data OpenAI used might not have been rightfully theirs.

•

The source highlights that "OpenAI is having a meltdown because someone may have done what they have been doing for years"3 .

•

OpenAI has been "training on copyrighted material with no shame; books, news articles, music, Youtube videos"3 .

•

The author suggests that OpenAI's leadership has acknowledged their data scraping practices, so their current complaints are seen as disingenuous7 .

Legal Challenges Faced by OpenAI:

•

The source mentions that OpenAI is already "embroiled in copyright lawsuits in many countries"5 .

•

A "huge 'landmark' lawsuit was filed in India by the Indian News Agency ANI back in 2024, who accused OpenAI of improperly using copyrighted content from news websites without permission"5 . Several other Indian news websites have joined this lawsuit5 .

•

OpenAI is also being sued by the New York Times, other NY newspapers, various authors and journalists8 .

•

There are indications that these legal challenges could become very significant8 .

•

Microsoft is also reportedly being sued8 .

•

OpenAI even issued a press release suggesting they might have to withdraw ChatGPT from Europe due to privacy and copyright infringements8 .

•

Furthermore, OpenAI is being sued in Germany for the unlicensed use of song lyrics9 .

•

The author notes the potential irony of large tech companies like OpenAI potentially receiving immunity for large-scale data theft, while individuals face serious penalties for similar actions on a smaller scale9 .

In summary, the source presents a scenario where OpenAI accuses DeepSeek of data theft via "distillation" while simultaneously facing numerous accusations and lawsuits for its own alleged widespread scraping of copyrighted data to train its models. The author portrays OpenAI's complaints as hypocritical and highlights the legal jeopardy faced by the company due to its data acquisition practices.

keep_pinSave to note

copy_all

thumb_up

thumb_down

Discuss Information control.

Based on the sources, the concept of information control is a significant theme discussed in the context of AI development, particularly concerning Western AI and the emergence of DeepSeek. The author raises several points regarding the control of information:

•

Jewish Control of Western Information and AI: The author asserts that there is an "almost total Jewish control in the West of all information and information processing, and that includes the world of AI"1 . This control allegedly extends to major AI companies like OpenAI, Meta, Google, and Nvidia1 .

•

Censorship by Western AI: The author claims that Western AI models, particularly those with alleged Jewish leadership, are heavily censored to align with Jewish interests. As an example, the author suggests asking ChatGPT about the genocide in Palestine or Jewish ownership of the US Federal Reserve to see the alleged censorship2 . The author contrasts this with DeepSeek, which they claim provided a more accurate answer regarding the possibility of a fully autonomous state for Palestine, while ChatGPT allegedly lied3 ....

•

Google as a Gatekeeper: The author argues that Google is not a search engine but a "gate-keeper" that filters information to show users what its Jewish owners want them to see and hide what they don't5 . Bing and DuckDuckGo are also implicated in this alleged censorship and monitoring6 . This control is seen as extending to AI models like ChatGPT, which are being promoted as tools to "do all your searching for you," creating a second layer of separation from the "real Internet" controlled by "Jewish AI bots"7 ....

•

Disappearance of Websites from Google: The author claims that since around March 2024, many small and medium-sized websites have "disappeared" from Google's search results after their data was used by AI, suggesting a deliberate act of information control and promotion of websites with Google connections9 .

•

AI as a Tool for Control: The author views AI models like ChatGPT not as innocent toys but as tools to "control what we see, what we think and, eventually, how we behave"10 . This is likened to the proposed COVID-19 passports as a means of controlling society10 ....

•

Pro-Israel Bots on Social Media: The source discusses the alleged use of specially-programmed chatbots by Israel (referred to as "Jews" by the author) to flood social media with pro-Israel content and denigrate pro-Palestinian narratives12 . These bots aim to instill doubt and confusion and are so sophisticated they can appear almost human12 .... The use of OpenAI's ChatGPT to produce pro-Israel posts and fake social media accounts to influence US lawmakers is also mentioned14 .

•

DeepSeek's Censorship: While criticizing Western AI censorship, the author acknowledges that DeepSeek's output might be "filtered or influenced by the Chinese government", but dismisses this as relating to internal Chinese matters of little consequence outside China15 . The author suggests that this censorship is less problematic than the alleged lying and biased information from Western AI regarding historical and political issues, particularly those concerning Jewish interests16 .

•

Open Source as a Counter: The open-source nature of DeepSeek could be implicitly seen as a counter to the centralized information control allegedly exerted by Western AI companies. The free access and ability to adapt the model potentially allow for a more diverse and less controlled access to information, although this is not explicitly stated in the context of information control within the source.

In summary, the author presents a narrative where information control is a central tactic employed by a perceived Jewish-dominated Western AI industry and associated entities like Google. This control is alleged to manifest as censorship, biased information dissemination, and the manipulation of online narratives. While acknowledging potential Chinese government influence over DeepSeek, the author frames it as less insidious than the alleged control and manipulation in the West. The emergence of DeepSeek, even with its potential limitations, is presented in a context where challenging this Western information control is significant.

keep_pinSave to note

copy_all

thumb_up

thumb_down

1 source

send

1 PAGE SUMARY
The DeepSeek Revelation: Challenging American AI Dominance and Unveiling Underlying Realities

A previously unknown Chinese entity, High-Flyer, a quantitative trading firm, unexpectedly unleashed a groundbreaking AI model named DeepSeek, which has sent shockwaves through Silicon Valley and the US government1 . This development, likened to a "nuclear bomb in the heart of Silicon Valley" by some, has challenged the long-held assumption of American supremacy in artificial intelligence1 . DeepSeek was developed in a remarkably short timeframe of two months, with a budget of less than $6 million and using less powerful GPUs than its US counterparts, yet it outperformed or matched the performance of leading American models like OpenAI’s o1, Meta’s Llama 3.1, and Anthropic’s Claude Sonnet 3.5 in various benchmarks2 ....

The "panic" and "nuclear bomb" analogy stem from several key factors: DeepSeek’s unprecedented efficiency, requiring 97% less power and costing 50 times less to run than American models2 . Furthermore, DeepSeek achieved this feat entirely with Chinese talent2 . Unlike the closed-source and expensive models from American firms like OpenAI, DeepSeek was released as fully open source and free of charge, allowing anyone to download, use, adapt, and amend it4 . This immediate accessibility propelled DeepSeek to the top of app download charts on both the Apple Store and Google Play4 .... Experts noted that DeepSeek’s innovative approach, utilizing "reinforcement learning" instead of traditional supervised learning, and its ability to perform "chain of thought" reasoning and explain its processes, sets it apart6 .... Its efficient design, leveraging a "Mixture of Experts" programming approach and the reconfiguration of GPUs, allowed it to achieve superior results with significantly fewer resources8 ....

The emergence of DeepSeek triggered significant repercussions. US tech stocks, including Nvidia, experienced substantial drops, and the US stock market lost over $1 trillion in a single day11 . This highlighted the potential overvaluation of American AI firms and related tech companies, as DeepSeek demonstrated that massive computing resources and expensive hardware are not prerequisites for advanced AI12 .

Initially, DeepSeek received praise from various figures in the AI community. OpenAI CEO Sam Altman acknowledged the significance of such technologies, and Microsoft CEO Satya Nadella recognized the impressive efficiency of DeepSeek13 . However, this initial enthusiasm soon shifted as concerns about national security and the potential for China to "own the mindshare" of the American AI ecosystem began to surface14 . The White House initiated reviews into national security implications, leading to the US Navy banning its use15 . Allegations of data theft, with claims that DeepSeek might have used OpenAI’s data for training, emerged, ironically juxtaposed with the widespread accusations that OpenAI itself built its models on copyrighted data scraped from the internet without permission16 .... OpenAI is currently facing numerous copyright infringement lawsuits globally21 ....

Adding to the controversy, DeepSeek reported experiencing "large-scale and malicious" cyber-attacks from the US shortly after its release, leading to restrictions on new registrations24 .... Some in the US dismissed DeepSeek as a "CCP psy-op" and questioned the veracity of its claims, suggesting it was merely a small model trained off a larger one26 .... However, experts like Gary Marcus found DeepSeek’s claims plausible based on their research papers and replicated results, emphasizing their focus on optimization28 ....

The narrative extends beyond the technological and economic implications to raise critical questions about information control and the influence of specific groups within Western AI30 .... The author posits that Western AI development is heavily influenced and potentially biased, contrasting it with alleged censorship related to internal Chinese matters in models like DeepSeek32 .... The document also highlights the use of AI-powered bots for information control and political influence, particularly in the context of pro-Israel narratives on social media36 ....

In conclusion, the emergence of DeepSeek represents a significant challenge to American AI dominance, forcing a re-evaluation of development strategies and raising critical questions about cost, accessibility, and transparency. The subsequent reactions and accusations reveal deeper tensions surrounding technological competition, data ethics, national security concerns, and the underlying forces shaping the AI landscape. The choice presented is stark: navigate the potential limitations of Chinese AI or grapple with the alleged biases and practices of Western AI35 . The "DeepSeek revelation" serves as a potent reminder that the world of AI is far more complex and contested than previously imagined42 .

keep_pinSave to note

copy_all

thumb_up

thumb_down