LyrArc Article Gist
Morning Star Firstlinks on AI investing which says that AI is not generating cash flow, it is burning it. It looks at the huge depreciation of data center technology over 5 years and that 25% returns on the AI center investments would not recover the cost of depreciation, let alone the whole investment. The depreciation for 2025 year is shown at $40 billion with $15-$20 billon in revenue in this hypothetical case. Revenue has to grow ten fold from what it is now to support the $40 billion in depreciation each year for these data chips.

LyrArc Article Gist
Issues raised by the huge mismatch between revenues and investment for AI. $400 billion estimated investment by 5 Tech firms in 2025 alone with revenue of about $40 billion and huge uncertainty about when AI will produce returns. Articles seen this week of November 17 in the WSJ and NYT on this issue, podcasts, discussions in other media outlets. Could this lead to a dot com bubble type economic crisis? Could that lead to a recession? Alongside these articles another article in the WSJ on Nov 17 shows the benefits small firms get by using AI, benefits which are on the fringes of their business, not essential but with some experimenting firm owners/managers able to tweak AI information for use in business. Nothing significant which firms will pay much money for. The uncertainty is a major factor. Should geopolitics trump all these concerns? Is the competition with China require this scale of investment, and is China following a more utilitarian approach as reported in a WSJ article this month, of investing in AI in a utilitarian way targeting its use in improving manufacturing, improving infrastructure, and not wildly throwing money at experimental uses that are unlikely to yield much result. In geopolitical sense would the country that not only promoted AI but used it efficiently and cost effectively, used it in ways that promote the overall public good, get the WIN. In short it behooves everyone of us to ask hard questions of AI, to dehype the hype, to look for the public good that comes out of this from it's efficient use. To ask the tough questions when $400 billion generates only $40 billion in 2025 and the $3 trillion planned investment over 5 years is half unfunded, is it going to crowd out energy needs for homes and business, push renewable energy targets back, crowd out essential investments in the crumbling aging infrastructure of the US and Europe, crowd out essential investments in education, healthcare, pharmaceuticals, and manufacturing, that hold better promise for our People. Will it also put retirees at risk when corporate bonds from retirees money fund the unfunded portion of AI? This means making the political dimension not about migration, settling the illegal migration issue that was meant to be settled a long time back, or about cultural issues that have little day to day impact on our lives which are about groceries, childcare, housing that are non ideological. Making the political dimension not about remote countries that one knows little about except when it affects public safety and health as with fentanyl. Capital allocation decisions to the vital needs of America can then be free of politically induced error, so that it can be subjected to the test of how best it serves the public interest and the people of the Nation. ...

LyrArc Article Gist
The uncertain returns from AI is not good for retirees whose funds are invested in AI funded by corporate bonds. With enormous needs for capital how about $3 trillion for Ai investment to 2028 , says WSJ, only $1.5 trillion can come from cash flows and $1.5 trillion has to come from elsewhere such as corporate bonds and annuities retirees purchase. This lowers credit ratings of these instruments as AI returns are uncertain. This is not a good situation for retirees in the US.

LyrArc Article Gist
How AI is costing America a lot in 2025-2026. $400 billion in investment in 2025 alone that is diverted from essential infrastructure repair and renovation in the US and for investing in making pharmaceuticals in the US, housing, and education, childcare, investing in the future in renewable energy.

LyrArc Article Gist
Prodigous investments in AI data centers is crowding out investment in essential infrastructure that would cut the cost of living in the US. Such as investment in pharmaceuticals in the US, investment in automobiles and rare earth processing, in housing and schools would reduce cost of living by bringing down prices and provide huge human returns for every dollar spent in addition to larger profits over a long period. Shown here is the AI data center for Microsoft in Atlanta. Microsoft has invested $34 billion the first fiscal quarter of 2025 alone, with similar investments by Amazon, Tesla, Google, and others for $400 billion capital allocation in 2026. Investments are also being crowded out in the replacing of the aging infrastructure of the US  of roads, rail, subways systems, transport systems, bridges, airports and ports. Some of these investments such as in ports and logistics are needed to make America a manufacturing and exporting nation. Economists loved to talk about crowding out of investment by the private sector when the government spending was significantly higher as during and after World War II. Today there is little talk about the massive misallocation of capital in the US economy. Where public infrastructure is ravaged by time and mismanagement as in New York political trends are calling for free public transport  and supported grocery stores in NYC, when the root cause the overall picture of the Nation's spending in rebuilding America is ignored or unaddressed, which would get to the root cause of the cost of living and quality of life issues that concern all the people of this Nation. ...

LyrArc Article Gist
Dangers of capital allocation to AI when other needs are great -housing and cost of living concerns. The prospects of boom bust cycle.

LyrArc Article Gist
Over the last 5 years 2021-2025 Americans who are more excited than concerned about AI has dropped from 18% to 10%, and Americans who are more concerned than excited has grown from 37% to 52%, in Pew Research surveys. Showing that self-interested tech companies such as Microsoft and OpenAI, Google are making loud claims for AI that do not reflect the views of the American people as a whole in 2025.

Americans by large margins in Pew Research believe AI will help in day to day tasks from weather forecasting to inventing new medicines. And by large margins of 40% Americans think AI will hurt ability to think creatively and form meaningful relationships and by 20% will hurt for making difficult decisions. This shows Americans -similar to people in China as reported- believe in using AI for ordinary day to day routine tasks, and are wary of AI and aware that AI's usefulness is limited to such routine tasks only.

LyrArc Article Gist
Rightly seen the reality in China, says Li Yuan in NYT is not about the tallest bridge at Guizhou, or the acceleration of an AGI race with China, but that enormous debt finances that bridge, pensioners live on $20 a month in that Chinese province. Doctors, lawyers and accountants provide the $1.50 ride share in that province suggesting that all is not right when it comes to the living standards of the people the ultimate test. When it comes to AGI and AI China is simply integrating it into processes and work for general efficiency, nothing strategic about it, merely routinely integrating a technology. There are deep structural flaws in China's development says Li Yuan, when there are enduring strengths in the US. It may be that Silicon Valley is more of a problem in the US as it seeks to divert more of the resources that should go into people benefitting infrastructure in the US. Ultimately the US must seeks its own path built on the expansion of frontiers in the US since 1787, followed by industrialization in the 1870's and again in 1940's and 1960's in a government of the people, for the people and by the people. ...

LyrArc Article Gist
Netherlands government takes back control of Dutch chipmaker Nexperia from China's Wingtech  using a court order in October 2025.

LyrArc Article Gist
The flaws of Wikipedia model in it's effort to be a purveyor of knowledge, after another failed attempt by Google, related to the basic flaws of their model. A flawed claim to be the sole purveyor of knowledge in the world forming a monopoly of their own, the risks to the Nation and the World.

LyrArc Article Gist
H1-B Visas and the Three Hundred Thousand Indian Engineers, the 3 million they would have trained locally in India to 2030, are a huge loss to India and India's dream of rapid modernization. The 3.3 million engineers in the 51 states of the Union to 2030, born in the USA, will also be lost to America's dream of re-modernization. It hurts the dreams of both nations for modernization of infrastructure and economic growth. 

LyrArc Article Gist
China's strategy appears to be to allow the transfer of TikTok ownership to an American consortium and focus it's efforts on renegotiating the 30% tariff on China and China's response of 10% on the US. China is making an effort to protect industries in China that have surplus production. China wants DJT to meet Xi for a Summit in Beijing instead of South Korea as the US has suggested. And Xi  offers to make a visit to Washington DC to improve relations. Scott Bessent and the USTR Greer in negotiations  with the Chinese trade negotiator Vice Premier of Economic Policy, He Lifeng, in Madrid, have come up with an agreement. The US COngress has passed legislation requiring the sale of TikTok to US owners because of sensitive information issues. China now reversed its policy in April when it opposed this as it now sees little advantage in that policy and can concentrate on better trade terms from DJT. A stumbling block is the flow of fentanyl for which DJT imposed 20% of the 30% tariff on China, the other 10% being standard for all countries. ...

LyrArc Article Gist
US China agreement on TikTok sale to American ownership as required by Congress, Scott Bessent announces that an agreement has been reached with China in negotiations for the transfer of ownership. China is concentrating its energies in negotiating tariffs that were kept at 30% by the US and 10% by China which hurt Chinese industry in industries where it has overcapacity.

LyrArc Article Gist
Lingling Wei on Chinese policy on trade with the US in the WSJ and China seeking a visit by DJT to Beijing instead of APEC side meeting in South Korea. A meeting in Washington DC is seen as risky after the Zelensky meeting with DJT took an unexpected turn, and the idea of meeting in Beijing gives more opportunity for getting an organized result and show China's standing in the world of nations. This happens after XI met Putin in Beijing on Victory Day celebrations for World War II where Russian and Chinese losses were far larger than European or US losses. China's huge losses in the millions have not received much attention in the US or Europe. This is also true for losses by the Philippines, Indonesia and India from decisions made during wartime by colonial powers and the Imperial Japanese Army. A meeting of Xi and DJT in Beijing from China's point of view may also show China is ready to work with the US in trade and the economy where it has huge interests in a stable transition to where Chinese industry does not overproduce what it cannot sell and seeks a diversified market shifting away from concentration in the US. Both Xi and DJT are playing to a domestic and international audience to show they are wise leaders willing to engage and at the same time protecting their national interests. The issues of support for Ukraine and fentanyl sources in China remain unresolved. ...

LyrArc Article Gist
WSJ with subscribers up to 4.1 million and other Dow Jones products, are key parts of revenue for Murdoch's business group in 2025. 85% of profits are derived from Dow Jones for News Corp. The Australian and British newspapers do not generate much profit and The Sun has seen years of losses. These newspapers gave Murdoch his start in Australia and UK, but have built up losses today.

LyrArc Article Gist
The campaign in German schools "Smart ohne phone" or "Smart without a phone"  for school children in 2025. One student at Dalton High School in Alsdorf, North Rhine Westphalia, Klara Ptak, is cited in this report by  Oliver Pieper in DW.com. Klara 17 years old says- "A total of 51 phones were seized — that's a fairly significant number considering we have 700 students. And you can really see a difference, especially in the younger kids. They used to stand around in a circle staring at their screens, and now they are playing football, badminton, or board games together. It's a dramatic shift." A survey by the Körber Foundation, and pollster Forsa, says that parents of children ages 12 to 18 complain that their children's time spent on social media is out of control and is their leading source of stress. The German National Academy of Sciences Leopoldina is for banning phones upto tenth grade, it wants to keep children under 13 away from the internet for social media completely. It is interesting to note that the younger student are fine with it and adapted quickly seeing the benefits. This shows that a new generation of student can be trained for a different lifestyle. Today in the Washington Post Linda McMahon and RFK Jr. team up on an article saying just this that time spent on screens lifestyles along with nutrition and food fail school children badly, reminding people about the saying in Latin since the 2nd Century by poet Juvenal -"a healthy mind in a healthy body." ( Latin -"mens sana in corpore sano").   ...

LyrArc Article Gist
Will AI reduce the curiosity that drives the creation of new knowledge and the curiosity for new discoveries that powered science and technology since 1600? Will it affect the human tendency and habit of asking questions, seeking out novel answers that is intrinsically human that AI cannot do? Scientific inventions that led to Europe leading the way and Asia falling behind after 1600 and new inventions taking place for 300 years with old theories discarded and new knowledge created are impossible under an AI arrangement. If AI existed in 1600 few new discoveries would have happened because they involve asking new questions and finding answers to these questions that take many years sometimes a lifetime of discovery and invention. Other weaknesses of AI are for example that it is fast but it cannot think- it is pieces of knowledge pinned together in different ways that come up from billions of pieces of information pieced together. It gives the appearance of thinking if one is not careful to look at it's process diligently. Its main source is using the public knowledge base built by Wikipedia, with other additions piled on top. Wikipedia may be wrong there are biases and tendencies to overlook facts in Wikipedia inherent in any knowledge exercise. These are then transferred to AI without anyone knowing about it openly, making it more dangerous in that it precludes creative thinking and finding solutions that never existed before to problems or questions, which can only be done by the human mind through its curiosity and stubborn dogged desire to find solutions to a problem.  ...