Artificial Intelligence, AI, such as the LLMs Grok, Claude, Chat-GPT, etc., are now advancing at an incredible rate.
I’ve seen several people post references to this, Something Big Is Happening, recently, and related AI-disruptions in business like Bro-Bots at the bottom of this Coffee and Covid blog post. Related to this with specific examples in video, Claude Opus 4.6: The Biggest AI Jump I’ve Covered–It’s Not Close. (Here’s What You Need to Know) Given the predicate of my first novel is AI gone wrong, and redeeming itself, commenting on developments is quite appropriate and in-theme for this blog.
Right off the top of my head, there are several things to address, aside from all the military applications (a huge topic on its own). We have the economics of the AI industry itself, confidentiality / privacy / privilege of the queries and output, marriage-market and social disruption from job losses, legal liability for decision / actions taken because of AI outputs, raising up the next generation of leaders who know enough on their own without AI to sanity-check AI output, socioeconomic disruption from job losses due to AI replacement, electrical power and cooling demands (more electricity and/or lots of water).
Military applications, and disruptions that come from new tech advances brought about by AI are huge and complex and speculative, worthy of long essays in their own rights.
1) Confidentiality. Unless you are running it on your own bare metal, then the interactions with AI, both inputs and outputs, and not confidential, proprietary, secure, or private from governmental or corporate search and discovery. This a a major problem for litigants or corporate researchers, or anyone who deals with privacy and private data that needs to remain private, including medical data. It’s as bad, possibly worse, than Microslop screenshotting you in Win11 every five seconds and uploading it to their servers. If you are running the prompts and managing the output on the cloud datacenter servers, then they know what you are thinking, how you are approaching the problem, what problems you are working on, and how you are likely to approach you solutions. How can you patent anything you come up with if they can beat you to the punch by having their AI automatically register any novel ideas it comes up with? Suddenly your new solution is their “prior art.” There are no laws to address this issue, just “terms of service.”
This fact also precludes using it on anything related to national security, unless it’s on special US Gov’t hardware and largely isolated from “casual” internet contact.
This represents a gaping hole for patient privacy, attorney-client privilege, potentially priest-confessor privilege, trade secrets, copywrite problems, and more. Once people realize the implications, they will either change how they use it significantly, or stop using it at all. The first company to address this adequately will reap a huge benefit in the marketplace.
2) Legal Liability. The issue of liability for decisions made or actions taken by AI that cause monetary damages or physical injury is not well established. The whole edifice may be one big court case from collapsing as “too risky.” The first question for a lot of people when something goes wrong is “who can I sue?” The AI companies will, of course, claim they are just a chat-bot and that we can’t take them seriously, and we should always get the advice of qualified professionals. But that sort of undercuts their value, if you have to constantly fact-check them with humans, doesn’t it? It also reduces their cost-reduction claims.
This is sort of like the self-driving cars and the “who is responsible for the crash?” problem, but writ large for every action and decision downstream from their output. It’s a legal nightmare, and again, there is no good legal framework to deal with this. Heck, there isn’t even really a bad legal framework, just a small handful of limited court-case precedence.
The AI companies are one big lawsuit loss away from bankruptcy until this aspect gets addressed at the national and international level in something like a comprehensive way.
3) Destroying the future top workers pipeline. By automating the low-level jobs, it’s gutting the “farm-team” of the next generation learning new skills and becoming established at the top of the field to be the top-tier robot-wranglers who sanity-check the AI output. Where would the next generation of experienced leaders come from? I mean, I know who the elites would /like/ to elevate to those positions, but I trust we all see the problem with putting baby-eating Satanists in charge of picking future “experts and leaders.” So, how do we train up the next generation of high-level, high-paying, highly-skilled top talent if there is no lower rung on the ladder for them to start the climb of experience? In order to sanity-check the AI output, they have to have been raised without the crutch of AI “help” crippling their brains with dependency.
It can be hard to know who will be a “top talent” in any particular field without having a large number of people in the talent pipeline. You can’t just say “well, the top 0.5% (“top” by what metric?) will get the hard-core education for field X, and they’ll be the future leaders…” at least, if you want them to actually be top talent, and not nepo-babies of the politically well-connected. The only way I know of to address this is to ban AI tools for the top 20% of the young, say everyone under 18, or maybe 21 years of age, so they learn to use their brains the old fashioned way before teaching them how to get the most out of AI. But that will be a hard sell to a population addicted to the quick fix of easy commands making them lazy.
A much more difficult problem to solve than the first two, because it really requires both a mass cultural shift and likely a legislative framework to address. It is not simply a knotty and landmine-strewn philosophical/legislative problem.
4) Profits and economics. None of the AI companies are profitable. In fact, they are burning cash at a huge and nearly Ponzi-scheme rate. They all hope someone figures out how to recover their investment costs, but it’s not known or present yet. The benefits are (potentially) huge and concentrated, but the corporations are not willing to pay what the AI really costs. We don’t have an economic model that makes this whole thing work, and continue to work for the long term. The rapid advance largely depends on the masses of money being poured into it, but the business model is still coming up woefully short. It’s too early to tell yet where and how it’ll really be a huge cost-saver, and what industries it’ll make bloom. Lots of speculation, hope, and hype, but little hard data yet; the target is moving too fast, but the costs are real. What happens if a corporation fires a lot of semi-specialized workers and automates via AI, but then the company providing the AI tools goes under? How do they recover? Can they recover if they don’t have any workers to recall to the office? Will that knowledge get lost forever?
And speaking of office, the commercial real estate situation is already nearly catastrophic with high vacancy rates and non-performing loans, so how will it deal with major corporate down-sizing as office-workers are replaced by AI? What will laid-off AI-replaced workers do for employment? At least imported workers still live and consume in the real economy around the company they work for, but unemployed locals will demand the imports get sent back home, reducing local demand even in service-sector jobs.
Perhaps this problem will be the solution to the other problems. If it’s too expensive when all the real current operating costs are factored in, it’ll self-limit to things where it’s really helpful and the cost-savings are huge in areas of massive economic friction, such as medicine, law, and fraud-detection (and prosecution).
5) Social adjustments. I have yet to see anyone seriously address how the already messed-up marriage-market (caused by feminism, socialism, dating apps, hypergamy, etc.) will deal with mass layoffs of women (who are heavily represented in the easily-automated workspace), fewer family-wage jobs for men that married women want their spouses to have, and some sort of UBI which will almost certainly result in further negative and escalating social / marriage population disruptions by creating an expanded and psychologically beat up parasite class of former professionals (or at least, a somewhat functionally employable, if now dated, skill-set). More women with degrees and debt in the workplace, all demanding “high value men” while offering little marital value themselves, will be competing for fewer jobs, while corporate interests will still be wanting to import lower-cost immigrants, will make for a socially toxic brew.
With fewer “respectable” or “high-status” white-collar jobs available to men that are family-wage, without a major social adjustment the women will be chasing ever-smaller pool of “quality” men. If recent history is any guide, their demands will go up with their debt load even as their qualifications go down. The men can’t compete with social-welfare benefits, and the legislatures do not appear interested in changing any of the welfare/income cliffs they’ve built into the programs that dissuade part-time or moderate-wage work. Women who can make more on the dole than by getting married will whine about how unfair it is, and maybe have kids out of wedlock, but won’t change. But single moms are not an optimal environment to raise sons in to be good workers and leaders in.
This will demand some serious thought, and likely a real propaganda campaign to alter the social stigmas and respect various choices get, as well as some serious legislative changes to address the underlying economic and social problems.
At the same time, a high unemployment rate will create even more pressure to send all foreign workers home. If the choice is kick out foreigners and pick fruit, or starve, at this point Americans are so demoralized and gas-lit I don’t know what most of them would do. However, enough will choose to send the foreigners packing that the Great Culling that is coming won’t get everyone.
It’s going to be a spicy few years as we adjust to these changes. Manufacturing jobs can’t be re-shored fast enough.
6) Power and cooling infrastructure. In many ways, this is the most straight-forward of the problems in my list; these AI data-centers need a massive build-out of reliable base-load power generation. Ideally nuclear or coal, but in any case it needs to happen fast, as their demand on the grid has made power-bill spike dramatically, and it will continue to get worse. In a best-case scenario, they’d find a way to put data-centers near places that need a lot of hot water (for heating, or whatever), so the waste heat can be dumped into water, piped to where the heat can be useful, rather than simply evaporated away. With any luck, this will kick-start a large push for new-generation nuke plants. The problem there, of course, is the stupidly long lead time, huge up-front costs, regulatory hurdles, and lack of standardization in parts, production, or training. Maybe this time they’ll get it done right. One way or another, though, this has to be addressed, and soon, given the huge push for EVs, shutting down coal and fossil-fuel base-load generation, and other stupid “green” decisions leftie politicians push.
This has many relatively simple and technologically feasible solutions, but they are politically somewhat more difficult. However, the pinch felt by voters over their power-bills may motivate some of them to see the light. However, if we don’t get the addressed, and soon, it will put a huge downward pressure on the economy by jacking up electricity prices to a pain-point nobody wants to face, and that will cost jobs. A LOT of jobs. Which will further reduce the tax base, while increasing the demand for social services, aka “the dole.”
Conclusion. There are some serious issues with the outward rippling effects of the AI revolution that people are not serious addressing yet, some of which are very complicated problems that our current industry and political leaders appear to be neither willing nor able to address in any serious way. But “no decision” is a choice in itself that has consequences. Interesting times, indeed.