AI Boom Carries Risks for US Economy

[marcus]

From https://finance.yahoo.com/economy/policy/articles/robert-rubin-warns-ai-boom-190255724.html
Title : Robert Rubin Warns AI Boom Carries Risks for US Economy, Markets
By Natalia Kniazhevich and Lisa Abramowicz , October 2026

... Rubin (Treasury chief in the late 1990s) said ... "Some of these very big AI companies have enormous commitments, and then there are a lot of suppliers, and a lot of suppliers have borrowed against those commitments," ... "What happens if they can't fulfill those commitments or all those borrowed against them? ...

He also pointed to inflation and a broader loss of confidence in the government's ability to address fiscal problems ...

Rubin pushed back against the argument that AI-driven productivity growth could allow the US to effectively grow its way out of its fiscal problems. While it's possible that a productivity acceleration boosts GDP gains, the benefits could be accompanied by substantial job losses, particularly among white-collar workers ...

Lawyers, accountants and television workers could all face displacement, he said, raising difficult questions about how people transition into new roles. The US currently lacks effective programs to deal with such disruptions, he said ...

He also questioned whether the massive investment in AI ($10.3 trillion from 2025 to 2032 - WSJ) will ultimately generate sufficient returns. "Will it pay off, will it not? ... "Some of this will turn out to be terrific and some companies will do great, and a bunch of other companies are going to do badly."

Rubin said his concerns extend beyond markets and the economy to the potential safety risks posed by increasingly powerful AI systems.

He said he sees AI safety and climate change as "2 massive existential risks" unlike anything humanity has previously faced ...

 

[marcus]

From https://finance.yahoo.com/technology/ai/articles/elon-musk-fires-back-broadcom-133013594.html
Title : Elon Musk Fires Back at Broadcom CEO Hock Tan's AI Investment Warning: 'No Crying in the Casino'
By Shomik Sen Bhattacharjee , Oct 2026

... Broadcom Inc. CEO Hock Tan told the panel that "A lot of investors have no clue what they're investing in, but set very extreme expectations, only to find out that it's almost like they are in a casino gambling".

Tan also questioned how efficiently companies are spending on AI ...

"People are just wasting money, spending tokens, trying to understand (AI)," Tan said, comparing companies to a child who burns through $10 of arcade tokens in minutes and learns nothing. (https://finance.yahoo.com/technology/ai/articles/elon-musk-takes-direct-swipe-024700291.html)

The concern has also put AI returns and capital availability in focus.

Tan's warning comes from one of the biggest beneficiaries of the AI infrastructure boom. Broadcom said fiscal 3rd-quarter AI semiconductor revenue jumped 221% year over year to $16.7 billion, while Tan expects 4th-quarter AI semiconductor revenue to reach $21.7 billion ...

The financial concern is widening. According to research published by the Brookings Institution in September, the U.S. AI infrastructure buildout is projected to reach $10.3 trillion through 2032. This investment is expected to average 3.63% of U.S. GDP annually, surpassing historical economic booms such as railroads and electrification, increasing pressure on companies to generate returns that justify enormous capital spending.

 

[pantaiema]

Well, everything in life, not just investment, carries risk.

The $10.3T figure published by the Brookings Institution, is not an annual budget commitment of hyper scalers or AI data centres companies, but it is Brookings's own assumption and projected estimate of the total cumulative spending added up over an eight-year period from 2025 through 2032.

In contrast, the total current annual spending remains below $1T. A significant portion of this immediate spending by hyperscalers and infrastructure companies is already locked in by contractual backlogs rather than pure speculation.

AI data centre infrastructure companies, hyperscalers CEOs are among the sharpest business leaders, the most intelligent people on earth that have a long standing reputation. They will not be so naive as to deploy capital without visible demand or expected revenue. In the future when the demand is easing they will simply cut their spending on AI Infrastructure.

 

[marcus]

... the most intelligent people on earth that have a long standing reputation ...

I am far from being a business specialist but from the little I know , from all these "most intelligent people on earth" lets take Elon Musk (the richest on earth) as an example : in my view his only really good & sussceful idea was the reusable rockets (part of SpaceX) . His other (probably also) good ideas (except Twitter ?) : Zip2 , Xcom , electrical vehicles/Tesla ,Twitter/X , OpenAI , XAi , Neuralink , Boring Company , didn't impress me yet . (https://en.wikipedia.org/wiki/Elon_Musk)

 

[marcus]

From https://www.techradar.com/pro/lack-...lobal-economy-if-the-ai-bubble-pops-imf-warns
Title : Lack of promised profit could tank the global economy if the AI bubble pops, IMF warns
By Craig Hale , Oct 2026

... Specifically, the IMF warned of 3 potential developments : a correction in AI-related investments and company valuations, an increase in government bond yields, or a major cyberattack targeting critical national infrastructure, including financial institutions ...

--------------------------------------

From https://www.techradar.com/pro/ai-ha...-250-million-jobs-is-the-only-way-to-solve-it
Title : AI has a $2.5 trillion problem and replacing 250 million jobs is the only way to solve it
By Benedict Collins , Oct 2026

... New analysis by Marin Ivezic, Founder of Applied Quantum, has found that the math really isn’t adding up. In order to justify the colossal investment, AI would need to generate approximately $2.5 trillion in annual revenue ...

Optimistic estimates put AI’s current revenues at between $150 billion and $250 billion per year, which is a multi-trillion shortfall ...

This is compounded by AI models that become cheaper to use but deliver far superior capabilities over previous models. GPT-4-class capability models for example have dropped in price 100-fold since 2022, but the costs borne by infrastructure providers have not dropped at all - and are rising instead - making it even harder for them to make a return on their initial investment ...