Control Panel: China's AI Moonshot

https://www.nytimes.com/2026/07/17/business/china-ai-moonshot-kimi.html

China’s Latest A.I. Breakthrough Threatens America’s Lead

China’s Moonshot AI unveiled a freely available artificial intelligence model that seemed to narrow the gap with cutting-edge offerings from U.S. tech companies.
By Meaghan Tobin and Cade Metz, The New York Times, July 17, 2026

Chinese start-up Moonshot AI released a new A.I. model that appeared to narrow the lead held by well-funded American competitors.

Moonshot said that the model, Kimi K3, was the world’s largest open-source A.I. system, allowing anyone to use, modify and build on it freely. The company said that Kimi K3 performed as well as leading models from OpenAI and Anthropic at some key tasks…

The release of a free, open-source Chinese A.I. model that could rival the performance of costly, computing-intensive systems from Silicon Valley’s best-funded companies has reignited concerns over whether the industry’s enormous spending spree on data centers is justified…

Many software developers and start-ups in Silicon Valley quickly adopted the model, largely because it was much cheaper than the leading American systems. It arrived just as U.S. businesses realized they had to find ways to cut down their A.I. spending… [end quote]

I wrote about this on Friday when the news broke.

This is a shot at the heart of the AI ecosystem whose gigantic spending and soaring stock prices are based on earnings that are 85% circular and only 15% from end-users.

It’s also a direct shot at Nvidia since the Chinese technology is not based on Nvidia chips that are embargoed from China.

I think this news is stunning. It could be a true trend changer. The S&P 500 index is 40% AI related companies. They are spending mind-boggling amounts to build data centers. Profits depend on end-users who will PAY for the AI service.

Will these paying customers justify the expenditures? Customers are already cutting back on AI spending because it costs too much. They are already shifting to the free Chinese model. The hyperscalers are counting on exponential adoption, ignoring the typical “S” shaped adoption of new technologies which tends to have a long growth phase followed by a plateau. This model doesn’t include disruptive competitors.

We won’t even begin to know the impact on the stock market until next week. And the ultimate impact will take longer to develop.

The SPX, NAZ and NVDA dropped on Friday but only at the level of noise. Next week will be interesting because Monday and Tuesday may see further drops and a wave of margin calls.

The Fear & Greed Index dropped slightly from Neutral to Fear. The trade dropped into risk-off but again that’s only one day and it wasn’t extreme. Bullish percent fell and VIX rose but it looks like noise, not signal.

The market is still in a bubble. One day won’t influence the CAPE which averages 10 years of inflation-adjusted P/E ratios.

The Chicago Fed’s National Financial Conditions Index (NFCI), which provides a comprehensive weekly update on U.S. financial conditions in money markets, debt and equity markets, and the traditional and “shadow” banking systems, shows that conditions are very loose and getting looser.

Valuations of the AI bubble are clearly driven by a typical momentum/ mania dynamic. Normally, valuations are strongly impacted by interest rates which are used to discount future earnings to a present value.

Rising interest rates impact the stock market in many ways. I wrote about the trend in long-term Treasury yields earlier this week. Long-duration yields impact business, mortgages and consumer loans more than the fed funds overnight rae.

The Treasury yield curve is rising at all durations but especially at longer durations. Bond holders want to be compensated for the risks of holding long-duration bonds. This is in addition to the risk of inflation since the long duration TIPS yields are rising along with the nominal Treasury yields.

I have added the price of gasoline and diesel to the Control Panel along with the price of oil. The war in Iran and Ukraine’s attacks on Russia have taken refinery capacity out of service. The price of refined products have a direct impact on the Macro economy. Oil, gasoline and diesel are all rising sharply since the so-called “Memorandum of Understanding” optimism has evaporated along with continuing attacks. So far, the “crack spread” (the ratio of the price of gasoline to oil) is holding steady despite the attacks on refineries. (@AlphaWolf no dirty-minded comments, ya’hear?)

USD continues in the channel that began a year ago after a sharp drop.

Gold and silver continue to fall. Copper is in a rising trend. Bitcoin continues to bounce along the floor.

The METAR for next week is potentially stormy. It depends upon how the bubble market responds to the news from China. The game of musical chairs is getting more challenging.

Wendy

13 Likes

Wendy,

It is probably worse than you imagine.

Iranian Drone engineers using it to guide drones.

For that matter,

Iranian
Korean
Ukrainian
Russian
Cartels
ISIS
and so on.

Chinese
Indian
Ukrainian
Russian
and others writing malware.

And worse

Chinese programmers using it to advance products and services for use inside and outside China. Followed closely by Indians, Thais, Filipinos, Mexicans, Chileans, Argentinians, and a bunch more.

It simply takes all of the guard rails off of AI and puts into the hands of everybody. With emphasis on EVERY.

Oh, and it is reasonable to expect that the Chinese Intelligence Agencies have a direct feed.

Deep Seek and I had long conversation about how the Ukrainian - Russian war ends and in the end it asked me, because I told it the Chinese were in the position of influence, what I would tell the Chinese leadership. I told it if the conversation hadn’t already been passed to Chinese intelligence community I would be disappointed in the Chinese leadership. It said, “Fair enough”.

Cheers
Qazulight

7 Likes

The economic impact of open source, “skim-free Free AI” on the overall economy is going to be significant, likely exceeding what the for-profit AI firms were banking on.

If you’re a LTB&H investor in the S&P 500, you’ll win either way.

But it’s still going to be a bad thing for anyone looking for a job.

Also, I’m not worried about excess data center capacity. Anything they don’t need for AI model development can be repurposed to a growing surveillance state to keep tabs on the restive unemployed population.

intercst

3 Likes