I could but I'd probably miss the angle that is helpful to you. Do you want to instead share what your pain points are, what are you trying to delegate to AI and what you aren't? Based on that I can expand my answer better with a thing or two you can try.
Raw materials in, lots of process steps in the middle with tolerances etc; some % yield of goods out.
When you work with LLMs, the "raw materials" is its first attempt to answer your prompt.
Mass manufacturing mirrors what we do with LLMs to knock their output into shape: grounding, automated quality checking tooling (eg. linters for code, grammar checkers for prose), and ultimately, as many turns as needdd to manually bang the output or work product into shape...
You should never ship your AI's first draft - that'd be slop. But you also shouldn't have to repeat yourself to get its output into a form you can use. If you know you're always going to have to tell it to change one specific thing, see if you can automate a process that tells it to change that thing for you, so you don't have to. Do this enough times and you've created a deterministic outcome at least at some level. And you're out of that loop of getting it to that level.
Tooling for this is limited today. You can provide instructions; but you're always at the mercy of labs to make models that follow instructions and at their mercy that they didn't put conflicting instructions in the system prompt.
Better to put in things like linters and grammar checkers - even going as far as putting in a fact checking process or for legal stuff, a citator.
Without these extras, AI output is no more real than some dream you had.
This doesn't ring true coming from the UK. I like in the US now and the healthcare is not comparable.
The UK health care system is an absolute joke, it's a running joke among a lot of the population. It's not free for a start, the taxes the population have to pay are eye watering. You have to wait weeks, if not months for a specialist, hospital beds are few, there is no accountability for regular fuck ups (I've been on the bad end of a horrific one).
The US system from what I have seen so far is a dream. It's cheaper and the service is top notch.
For context I've moved from the Midlands in the UK to the Research Triangle in North Carolina.
A thing people in the US have trouble keeping a bead on is that net compensation (after taxes) in Europe tends to be sharply lower in the US, so that white collar professionals with employer-provided insurance are generally coming out significantly ahead of where people in the NHS are, financially.
Incomes are sharply more unequal. Yes, tech professionals earn more in the USA. Plenty of people flip burgers or cut lawns, and 25 million or so can't afford health insurance.
If you take any kind of look at disparities in health outcomes, it becomes immediately apparent who is dragging down the mortality statistics. This is not something espresso-drinking researchers at WHO dreamed up to fight a culture war.
I'm not making an assessment of which economy or system is better. I'm saying that the modal American HN commenter would probably be significantly worse off financially in the UK NHS than on their own insurance, even factoring in how expensive insurance is.
That's true, and it is, but people's inability to be clear-eyed about the politics and the incentives are a big part of why we're locked in this situation. Proponents of payer-side reform are, from the perspective of a huge number of Americans, selling a bill of goods. Ordinary PMC suburban households will very likely be materially worse off.
We have done a number of things about it, most of which are more expensive and less effective than the very simple option of just adding them to Medicare, but expecting people to change from what they know and like to benefit less than 10% of the population is not a winning strategy.
Medicare shares almost the problems of the private system and adds others. The current Medicare scheme, of single-payer health care for the cohort most in need of health services, makes a lot of sense.
Medicare does have many problems and arguably isn't cheaper than our current private systems, but I would say a large portion of the uninsured also are part of the cohort most in need of health services (or they're young and healthy and cheap to insure/treat and largely irrelevant), which is why I think it would make sense to just roll them into some public plan instead of using EMTALA as a catastrophic insurer of last resort.
I agree and am generally sanguine with any service we want to roll out to the Medicaid cohort, but recoil from the idea of standardizing all health care in the US on Medicare, not because I'd be worse off (though, like every PMC household in the US, I probably would), but because Medicare is not good, and we'd basically be enshrining all the failures of our system in law and hiding the costs in the tax code.
Mate, I don't know how long you have been here but you have another think coming.
NHS services are free at the point of care. I worked in the NHS for a good few years. Never swiped anybody's credit card.
In the USA, calling an ambulance will set you back more than $1,000. No insurance? Better not give birth. You'll need 20 grand. Maybe 50 grand for a C-section.
My first child was born in the USA. My wife was lying on a gurney in a corridor for hours because all the operating rooms were full. By the time we got into the OR she was fully dilated and it was too late to get an epidural. It was complete chaos.
Our second child was born at Kings College Hospital in Lambeth, in a bed, with a midwife, listening to music. No jacuzzi required.
The US spends literally twice as much as the UK on healthcare per capita. So if by cheaper you mean literally 100% more expensive, then I guess you're on to something.
I'm no Trump fan, I have to say that many of these evidences have been debunked and quantifiably false. The raw and hours upon hours of Capitol Hill footage odeant show what the media historically lies about for example.
Have you thought that there was a massive physical infrastructure left behind by the original railroad builders, all compatible with future vehicles? Other companies were able to buy the railroads for low prices and use.
Large Language Models change their power consumption requirements monthly, the hardware required to run them is replaced at a rapid rate too. If it were to stop tomorrow, what would you be left with? Out of date hardware, massively wasted power, and a gigantic hole in your wallet.
You could argue you have the blueprints for LLM building, known solutions, and it could all be rebuilt. The thing is, would you want to rebuild, and invest so much again for arguably little actual, tangible output? There isn't anything you can reuse, like others that came after could reuse the railroads.
I've been using GrapheneOS for years, I can't go back to another OS due to its ease of use, speed, and awesome features baked into my day to day use now.
There is one banking app that stopped working, and you know what? I dont use it now. I'm not about to let a bank dictate how I use my most personal device. I use a desktop if I need to access that info, and it forces me to be deliberate about it too.
reply