I agree with you. This generation of LLMs is on track to automate knowledge work.
For the US, if we had strong unions, those gains could be absorbed by the workers to make our jobs easier. But instead we have at-will employment and shareholder primacy. That was fine while we held value in the job market, but as that value is whittled away by AI, employers are incentivized to pocket the gains by cutting workers (or pay).
I haven't seen signs that the US politically has the will to use AI to raise the average standard of living. For example, the US never got data protections on par with GDPR, preferring to be business friendly. If I had to guess, I would expect socialist countries to adapt more comfortably to the post-AI era. If heavy regulation is on the table, we have options like restricting the role or intelligence of AI used in the workplace. Or UBI further down the road.
No, it will not affect you. The above post is mostly correct, but it's misleading because it's from the perspective of a trader trying to make short term profits. But most of us are considered investors, we periodically buy/sell low volume of liquid ETF/funds/stocks and hold our positions for years. Market makers do collect a tiny premium for every trade, but it's irrelevant for the time horizon that investors are concerned with.
Since this is hackernews, I'd imagine there are a decent amount of people with low float tech stocks. Those things can be impressively squirrely. I've moved a billion dollar company 5% with a $100,000 order by being a klutz.
Granted, these situations are usually in times of market stress, but these are times when for better or for worse people do need to raise personal cash on occasion.
> but what was the benefit of translating from machine code to assembly instructions
In order to read it, I suppose? There's no reason for memorizing binary encoding schemes. I mean, you will learn that 0x90 is NOP on x86 but that doesn't help you a whole lot.
x86 performance is usually better with smaller instructions, which can be accomplished by writing code that doesn't require prefex bytes and also using certain instructions that sometimes have alternative shorter encodings for specific registers.
You can test this hypothesis by questioning GPT-4 until you hit a detail that it doesn't know. It will say something like "Without more context or a clearer reference to this detail in my training data, I can't provide a definitive answer.". Then you ask it to speculate about what the answer might be.
On Reddit, I've heard from people in the games industry that concept artists are being replaced by AI like Midjourney. Concept art can be generated via a prompt by the game designers and art directors.
I tried asking GPT-4 a math question and it does do math:
Me: "What (8 * 25) / 14 + 4 - 2?"
AI: "Let's break down the expression and solve it step by step:
Multiply 8 and 25: 8 * 25 = 200
Divide the result by 14: 200 / 14 ≈ 14.2857 (rounded to four decimal places)
Add 4 to the result: 14.2857 + 4 = 18.2857
Subtract 2 from the result: 18.2857 - 2 = 16.2857 (rounded to four decimal places)
So, the result of the expression (8 * 25) / 14 + 4 - 2 is approximately 16.2857."
For the US, if we had strong unions, those gains could be absorbed by the workers to make our jobs easier. But instead we have at-will employment and shareholder primacy. That was fine while we held value in the job market, but as that value is whittled away by AI, employers are incentivized to pocket the gains by cutting workers (or pay).
I haven't seen signs that the US politically has the will to use AI to raise the average standard of living. For example, the US never got data protections on par with GDPR, preferring to be business friendly. If I had to guess, I would expect socialist countries to adapt more comfortably to the post-AI era. If heavy regulation is on the table, we have options like restricting the role or intelligence of AI used in the workplace. Or UBI further down the road.