How Silicon Valley is prepping for War

As frontier models join forces with the Pentagon, corporations and the state are becoming too close.

Insider reports of OpenAI’s Orion and Gemini 2 models yet to be publically released suggest that Generative AI and LLMs might be reaching a point of diminishing returns, so where will the new profits and revenue come from for the Generative AI machine inside of BigTech and the research labs and model builders being funded by them? Look to the State.

I’m noticing an increasingly dystopian trend in Silicon Valley. AI research and BigTech are supporting national security initiatives that will lead to the militarization of AI and more dystopian applications of Generative AI like automated weapons. This has a long history in companies such as Google. With the inauguration of Donald Trump in 2025, this will take on a new form.

In 2018, Google faced significant internal protests regarding its involvement in Project Maven, a U.S. Department of Defense initiative aimed at using artificial intelligence to analyze drone footage and in machine learning assisted targeting. The project raised ethical concerns among employees, leading to widespread dissent and public outcry. Yet Google Maven set a precedent of the normalization of this trend and six years later significant BigTech software, data analytics and machine learning systems have been used inside of tools used by the Israeli Defense Forces (IDF) to slaughter innocent people in Gaza, among many other applications. The Ukraine war has supercharged drone development, with the U.S. leading the charge.

It’s not a vibe change in AI policy, Billionaires and BigTech have actually always been pushing for this and in 2024 they have won. They want to normalize AI in war, because it is profitable. Palantir a data analytics firm is the poster-child for this, and their stock PLTR 0.00%↑ is up 250% so far in 2024, mostly simply due to revenue from the U.S. Military. It’s extremely profitable to align yourself with the military industrial complex and the Pentagon.

This all started around September, 2017 and picked up speed in October seven years ago with Google Maven. As Amazon is set to fund Anthropic even more soon, Palantir and Anthropic are partnering with Amazon Web Services (AWS)¹, to make Anthropic's Claude models available to U.S. intelligence and defense agencies, the companies announced last week. Anthropic was supposed to be the more ethical version of OpenAI. It’s not about the research you do in the AI, trust and safety world, it’s what your company does with its tech that’s the important thing.

Anthropic funded primarily by Amazon and Google, are following in the footsteps of the company they are a splinter startup of, OpenAI. It was in January, 2024 that OpenAI had already announced their decision to go down this path working directly with the Pentagon. In late 2024, the U.S. even routinely gives out orders to foreign firms to suppress China’s ability to be independent in the semiconductor industry and AI. The U.S. has reportedly ordered TSMC to halt shipments of advanced chips to Chinese customers that are often used in artificial intelligence applications.

Back at the end of 2017, employees at Google articulated their stance through an open letter addressed to CEO Sundar Pichai, stating that Google should not engage in warfare technology. In 2024 and these days currently, all employee activism is shut-down well before it can escallate inside of the likes of Google, Microsoft or Amazon. BigTech and Cloud computing leaders work with the Pentagon increasingly and, not just in cybersecurity. Like I warned the people at the Pentagon involved in project Maven all those years ago, this will escallate. Now as those people retired a while back, it already has escalated and will continue unabated.

BigTech goes full on DARPA mode

In 2024 with the authoritarian MAGA version of American democracy, U.S. exceptionalism is the rule and it’s now a cool time to be advocates of AI Supremacy for the American military complex and supporting Venture Capital infrastructure, which is making companies and tycoons extremely profitable due to to catering to this currently dominant sentiment and AI policy approach by the likes of Palantir, Anduril, Scale AI, Shield AI, Skydio and OpenAI, among many others.

Silicon Valley has decided to side with DARPA and Nationalism here too is on the rise. DARPA stands for the Defense Advanced Research Projects Agency (DARPA). Who knows what OpenAI’s and Anthoropic’s models will soon be used for inside the Pentagon and more classified projects. Anthropic deciding to take this path is a major ethical breach of their alignment policy. It shows you where America is heading if the top AI research labs have already been bought out.

This AI arms race is after all for me personally the reason which I started this Newsletter and called it “AI Supremacy”, which sounds like a very odd name for a Newsletter to use and many of my readers are often a bit baffled about what its significance might pertain to. 2025 marks the beginning of a time of scale and integration of AI capabilities into National security and foreign interference systems. This systems will only become more sophisticated, nefarious and effective. While the rest of the world scrambles to keep up and be part of this Generative AI future even the likes of Saudi Arabia is planning a new artificial intelligence project with backing of as much as $100 billion as it seeks to develop a technological hub to rival the neighboring United Arab Emirates (the UAE). Like innovation around Quantum computing, this isn’t just about science. It’s about nationalism and building entrenched economic moats.

Let’s turn our focus to China and the supposed National security reasons why this is happening. The collapse of China’s economic growth means it has major structural reasons and was well explained in a recent talk given at the World Knowledge Forum by Michael Beckley. Michael Beckley is the Director of the Asia Program at the Foreign Policy Research Institute, an Associate Professor of Political Science at Tufts University. Listen or watch the YouTube talk here:

What follow is a Customized NotebookLM Audio Overview on this talk: (access it here). This Audio overview is 23 minutes and 29 seconds long:

Listen to the AI Podcast 🎧

U.S. Elites and the Pentagon insist China is a Peaking Power

The idea here is that a China in decline is a more dangerous version of China geopolitically. The Biden administration has followed the whims of people like Eric Schmidt in their national security and AI policy who coincidentally also has a war drone startup. Silicon Valley has totally infiltrated Washington, the separation of BigTech and the State is more in name now rather than in rule of law or actuality. The Department of Defense and U.S. intelligence communities have awarded $53 billion in contracts to major tech firms over a three-year period. As incentives commingle, the institutions of democracy, capitalism and rule of law are starting to show signs of breaking down.

Silicon Valley, Tycoons and Washington are Aligned

In 2025 the likes of personality such as Elon Musk and Sam Altman will have even more unprecedented power. Tycoons by any other name. The tycoons are taking over. Meanwhile employee activism at BigTech have been significantly and ruthlessly been dealt with. These aren’t the symptoms of a benign pivot to National security. I’m noticing a lot of AI policy types doing what amounts to PR for the companies they work at.

The Biden Administration ushered in an era where the U.S. can signal what companies like ASML in the Netherlands, Nvidia - which has significant business in China, or TSMC which it gave money to in Arizona can do, and this is what I mean about U.S. exceptionalism. ASML and TSMC are’t even American companies. The Department of Commerce sent a letter to TSMC imposing export restrictions on certain sophisticated chips, of 7 nanometer or more advanced designs, destined for China that power AI accelerator and graphics processing units (GPU). This is not normal.

Back in 2017, Google’s human resources and ethics compliance teams dismissed those concerns of employees, but here we are just seven years later. BigTech is a war machine and will become implicated in civilian deaths like it is in Gaza and Lebanon. U.S. tech used by the IDF goes contrary to all that machine learning scientists and AI policy people stand for, but nothing has been said on the U.S. side. Even Silicon Valley tech news coverage has not covered the suppressed activism inside of BigTech itself. Something is wrong.

Recently Donald Trump claimed that Taiwan is stealing US semiconductor technology. Pre election on a Joe Rogan podcast, the future POTUS accused Taiwan of stealing America’s chip industry and said he’d put a tariff on Taiwan’s chips. In reality of course, the U.S. begged TSMC to set up shop in Arizona, where TSMC was awarded $11.6 billion in CHIPS Act funding, made up of $6.6 billion in grants and an additional $5 billion in loans. An Act that gives funding to Intel ($8.5 Billion), an incompetent company in decline if I ever saw one. What is going on here?

The rest of the article continues for paid subscribers, go to AI-supremacy.com for further insights.