- cross-posted to:
- techtakes@awful.systems
- cross-posted to:
- techtakes@awful.systems
OpenAI, a non-profit AI company that will lose anywhere from $4 billion to $5 billion this year, will at some point in the next six or so months convert into a for-profit AI company, at which point it will continue to lose money in exactly the same way. Shortly after this news broke, Chief Technology Officer Mira Murati resigned, followed by Chief Research Officer Bob McGrew and VP of Research, Post Training Barret Zoph, leaving OpenAI with exactly three of its eleven cofounders remaining.
This coincides suspiciously with OpenAI’s increasingly-absurd fundraising efforts, where (as I predicted in late July) OpenAI has raised the largest venture-backed fundraise of all time $6.6 billion— at a valuation of $157 billion.
Microsoft and Nvidia are investing but they are both positioned to gain a return on investment by cloud hosting and hardware sales respectively.
Apple dropping out indicates they saw behind the curtain, did the math and realized they can create their own platform for less.
There is no path to profitability for them. Unlike other recent billion dollar tech companies there are no human workers to squeeze. Hardware doesn’t work for less than minimum wattage.
There is no path to profitability for them. Unlike other recent billion dollar tech companies there are no human workers to squeeze. Hardware doesn’t work for less than minimum wattage.
Kind of funny to see them Trying to capitalism squeeze physical laws And fail.
Good summary. Basically microsoft and nvidia are the ones selling shovels in this gold rush. (Especially the greens)
ChatGPT Plus is likely, for many people, a “lifestyle product.” And the problem is that, when people lose their jobs or inflation hikes, these products are the first to get slashed from the household budget.
So I have a slightly different experience here. When I lost my job recently I actually ended up signing up for ChatGPT plus. I abused the ever living hell out of 4o to crank out tailored cover letters and matching resumes. I was able to roughly triple my job search productivity until I got a job three months later.
Was it worth it? For that timeframe (3 months, $60) hell yeah since the mental labor of handcrafting cover letters for each job listing is extremely taxing and takes some of the awfulness out of the entire job hunt.
Can you share a bit about how you used it? I’ve used Copilot a bit to try the same thing, but it makes so many errors that I spend too much time editing and fixing them. Also, after running quite a few cover letters, I found that the text was repetitive and unnatural in a way that made it really obvious that it was an LLM writing the letter and not a person.
It helps when you’ve fed a few that you’ve made by hand to start in the same thread for it to use as an example (and format), along with one of your resumes. Then copy and paste the job description and have it generate a cover letter for you.
Keep it all in one massive thread, if it makes a mistake, correct it and tell it to apply those changes to future ones as well (in my case it kept saying I had over ten years of experience when in actually I just had ten, so I had to correct that behavior).
Since it’s an AI it will sometimes hallucinate, this usually happens if there are terms in the job description that aren’t in your resume… either have it regenerate (if it will take more than a few minutes to edit) or strip out the offending sentences. Some will need very little editing, especially if the job description closely aligns with what you have on your resume.
Oh and be polite because it now knows all your skills and can probably murder you in your sleep lol
Thanks, I’m guessing the benefit of subscribing is to create that persistent relationship. The free version from MS that I’m using times out after a while. I definitely get the problem of it making up experience for me when it encounters something in a job description that isn’t referenced anywhere in my info. Honestly, I’d probably get more interviews if I just let it make up stuff, but I’m guessing that might become a problem for me later. :)
I feel like if ChatGPT were the only LLM on the market, they’d have a real path to profitability, but it’s not even the best LLM on the market. And the open source models are nearly as good, meaning the vast majority of people who need an LLM can run it on their own hardware.
It’s kind of like trying to make a profitable business out of offering a special sauce that isn’t as good as your competitors sauce, and is barely better than the free sauce from Taco Bell. Oh and it costs you millions of dollars to produce a single bottle.
In other riveting news, water is wet.