The problem with new technologies so irresistible that they invite massive investment is that they may never pay back that investment. When I began in the consulting business in 1972 with The Boston Consulting Group, I was exposed to the common sense principle that it is cash that counts, not profits. When one considers public policy, quite often governments look at “cash traps,” businesses that will never produce as much cash as their overall investment requires. Semiconductors are one such business. Their investments are actually subsidies, typically by Asian governments, who see them as strategic and as critical inputs to other businesses that do generate positive cash flow. The bailout of the American auto industry in 2008/9 is a good example of investment that kept the United States in the game but now China is out-investing us without caring if they get a return. I’m not sure whether AI will ever make return its investment unless it becomes a valuable input to other business areas such as health care.
You are correct. Some years back, anyway, it was the case that for all airlines since the inception of commercial aviation in the U.S., the industry had lost money; I mean, net, overall, billions and billions in, billions and billions out: no net profit, only net loss. The same was true for the railroad industry which had a huge net operating loss since its inception in 1830--all this despite vast land grants and support by the federal government and various states.
The railroads and airlines created great wealth. They invited gargantuan investments. They created millionaires and billionaires, and benefitted the public, as well. but they lost money.
The currently developing technologies are likely to provide more examples. Losing money, though, is not their chief fault. The chief fault is that they will result in a techno-feudalism that will make most of us poorer, and benefit mostly the super-rich.
Trump announced yesterday that his great plan for AI is to not burden it with regulations, other than to make sure it isn’t “woke.” He described it as “a beautiful baby that’s born.” “We have to grow that baby and let that baby thrive. We can’t stop it. We can’t stop it with politics, we can’t stop it with foolish rules.”
What could possibly go wrong?
I realize that Trump & Chumps couldn’t care less about the planet all our lives depend on, but one of my big concerns with AI is its impact on our already stressed-out environment. “Rapid development and deployment of powerful generative AI models comes with environmental consequences, including increased electricity demand and water consumption.”
One of the chief inputs for AI is electric power. The Trump war on renewables will make electricity more expensive and more polluting, and it will destroy America's dominance in a field that likely will control the fate of the world. Already China is installing more renewable energy than the rest of the world combined, and their AI runs cheaper and more efficiently (though, at the moment, it seems, not quite as effectively).
At the same time, it seems to me that the potential efficiencies and speed of quantum computing may so drastically decrease costs that my concerns over electricity will be obviated.
Which way would you be betting?
And what is the future of cryptocurrency in a world of near-infinite computing (code-cracking) power? Seems dismal to me. What do you think?
The problem with new technologies so irresistible that they invite massive investment is that they may never pay back that investment. When I began in the consulting business in 1972 with The Boston Consulting Group, I was exposed to the common sense principle that it is cash that counts, not profits. When one considers public policy, quite often governments look at “cash traps,” businesses that will never produce as much cash as their overall investment requires. Semiconductors are one such business. Their investments are actually subsidies, typically by Asian governments, who see them as strategic and as critical inputs to other businesses that do generate positive cash flow. The bailout of the American auto industry in 2008/9 is a good example of investment that kept the United States in the game but now China is out-investing us without caring if they get a return. I’m not sure whether AI will ever make return its investment unless it becomes a valuable input to other business areas such as health care.
Erich,
You are correct. Some years back, anyway, it was the case that for all airlines since the inception of commercial aviation in the U.S., the industry had lost money; I mean, net, overall, billions and billions in, billions and billions out: no net profit, only net loss. The same was true for the railroad industry which had a huge net operating loss since its inception in 1830--all this despite vast land grants and support by the federal government and various states.
The railroads and airlines created great wealth. They invited gargantuan investments. They created millionaires and billionaires, and benefitted the public, as well. but they lost money.
The currently developing technologies are likely to provide more examples. Losing money, though, is not their chief fault. The chief fault is that they will result in a techno-feudalism that will make most of us poorer, and benefit mostly the super-rich.
Trump announced yesterday that his great plan for AI is to not burden it with regulations, other than to make sure it isn’t “woke.” He described it as “a beautiful baby that’s born.” “We have to grow that baby and let that baby thrive. We can’t stop it. We can’t stop it with politics, we can’t stop it with foolish rules.”
What could possibly go wrong?
I realize that Trump & Chumps couldn’t care less about the planet all our lives depend on, but one of my big concerns with AI is its impact on our already stressed-out environment. “Rapid development and deployment of powerful generative AI models comes with environmental consequences, including increased electricity demand and water consumption.”
https://news.mit.edu/2025/explained-generative-ai-environmental-impact-0117
Great guest post. Thanks!
One of the chief inputs for AI is electric power. The Trump war on renewables will make electricity more expensive and more polluting, and it will destroy America's dominance in a field that likely will control the fate of the world. Already China is installing more renewable energy than the rest of the world combined, and their AI runs cheaper and more efficiently (though, at the moment, it seems, not quite as effectively).
At the same time, it seems to me that the potential efficiencies and speed of quantum computing may so drastically decrease costs that my concerns over electricity will be obviated.
Which way would you be betting?
And what is the future of cryptocurrency in a world of near-infinite computing (code-cracking) power? Seems dismal to me. What do you think?