In a gold rush it is better, on average, to sell the shovels then to mine the gold.
Because its a scam.
A scam that will inevitably end humanity on a global scale…… probably.
Because nobody but nvidia profits from LLMs.
This is exactly the question i ask “investment advisor” services when they call asking if i want help managing investments.
For you to have a system that works well enough for me to be interested , there’d be no need for you to spend money on call centers full of people ringing up strangers for money, you could take all of that money and invest it in the plan you are trying to sell me.
I’ve always thought that as well. Surely it wouldn’t take them long to become ludicrously wealthy with their top investing acumen.
Why does it cost so much to engineer new models? I thought AI can replace software engineers now, shouldn’t the development cost of new models decrease over time?
Joking aside - I think the whole “engineer time is expensive, computer time is cheap” approach stopped being true half a decade or so before LLMs were a thing. It was kind of self defeating - you’d think companies were trying to increase their usage of computing resources due to some misconception that it’d automatically decrease work humans have to do, saving them money in the long run.
I don’t have the numbers, but I wouldn’t be surprised if software/hardware costs exceeded engineering costs for about a decade now, with all these inefficient-on-purpose frameworks and with everything being on the cloud and with using various services for the sake of using a service.
You can’t avoid the compute requirements
They’re working on it. They just need a little more of your internal documents and source code.
They’ll announce soon that personal ai assistants require a blood sacrifice to run, on top of the monthly and metered fees.
Because it is a scam. Just like the people selling, instead of using:
- healing crystals
- essential oils
- blockchain
- penis-enlargement pills
- JIRA
Jira is awful 😞😔
I mean yes, I hate the manager who set up JIRA that way, but JIRA is marketed to and designed for managers to over-complicate it. It is not designed as a tool for engineers to track their work. It is designed for managers to mis-manage teams, because that sells at volume.
I also have other reasons for hating JIRA:
- no ticket hierarchy
- no multiple owners
- slow. so very slow
- sometimes keyboard shortcuts work, sometimes they don’t. There is no apparently pattern.
I mean, it kinda fits with the rest of that category, other than I’ve never been harassed about healing crystals.
Q: Couldn’t this JIRA ticket just be an email?
A: No! Because JIRA will send you like, 10 emails.
LOL Love the (accurate) JIRA inclusion there
Jira, lol.
Been shouting this from the rooftops since the beginning. If it was half as good as claimed, I would NOT have access to it.
AI has sequestered the inventory and future production of memory starting at DDR5 and up. Meta has made an abomination “adapter” to use ddr4 in place of ddr5. They supposedly have backorders to 2030.
The idiots declared they had discovered fire but instead were merely rubbing two rocks together with such force to make sparks and smoke. A lot of people know they called it to early and are now trying to brute force create AGI.
AI capital investment is rumored to be over a trillion USD for just 2016. Microsoft has revenue at just under 300 billion. Google’s revenue is around 350 billion. Both of these companies are going to drown when someone finally calls it.
If I recall correctly, shovel companies made a lot of money while only a few of their shovels actually struck gold.
Recent scholarship confirms that merchants made far more money than miners during the gold rush.
The wealthiest man in California during the early years of the rush was Samuel Brannan, a tireless self-promoter, shopkeeper and newspaper publisher. Brannan opened the first supply stores in Sacramento, Coloma, and other spots in the goldfields. Just as the rush began, he purchased all the prospecting supplies available in San Francisco and resold them at a substantial profit.
There’s a lot of reasons;
- The people who run these businesses have grown incredibly wealthy by taking investment to grow. They do not need to be profitable
- Selling inference is massively profitable. They lose money on model training
- If you’re suggesting they build a “product” that is profitable, I would argue that’s a bad idea- it’s a high competition space. The people selling AI are in a much better position
Better question is if you can really build an app as easily as writing a prompt, why would anyone pay you for your app?
Somehow AI bros like to pretend that AI is some secret lifehack despite it being one of the most mainstream things in human history.
It still takes time and effort to be a project manager for app developer agents. Might as well just pay for it and get stuff done today. Instead of spending time on software maintenance, you could just pay for someone else to deal with all that hassle.
That’s the neat part.
I’d wager it’s simply not possible to be profitable using AI for most use-cases. You either burn way too many tokens on your queries or risk your business falling victim to an under-tokenized AI hallucination.
Or any AI hallucination
I love the assumption that if you spend enough tokens, it won’t hallucinate, like it isn’t a mathematical certainty or anything.
I agree with you, but I liken to to how companies tend to replace competence with bureaucracy. With enough layers, a significant amount of mistakes get caught. It just slows everything down and costs a bunch of money. There are still problems that get through, just less of them.
Right like. The more tokens you spend, the more hallucinations you get.
You can cut hallucinations by 90% just by prompting the LLM to.not guess API and function names and to provide it with full docs or MCP to find docs.
Yeah but it’s not guessing in the first place. It has zero reasoning skills. It just outputs the most likely next word. Even when you ‘tell it not to’
I know I am getting down voted for this because people haven’t tried it. And let me differ, it has reasoning skills, it now gets beyond predicting the next word. And even reducing it to next word is some kind of primitive reasoning.
By experience, by telling the AI to not guess APIs and function names and instead provide it with a way of looking through docs, and inspecting the code and other tools to get it do some tests, the hallucinations disappear. I’ve been able to have better results with this setup.
This is very different than the “no mistakes” joke.
“it has reasoning skills” Nope. It has larger context windows and someone coded in the ability to reprompt itself for “thinking” steps that are literally just a new prompt with the same context that gets fed back in which merely increases the chances of it guessing the next output token correctly.
It still doesn’t “reason”. It still doesn’t know what the words mean. The modified algorithms are just attempting to generate the corrective prompts for you. If you’ve ever messed with them long enough, you’d understand that these, “no wait fix it” prompts have rapidly diminishing returns at the best of times. Their “thinking” is a bandaid, and a piss poor one at that.
I know I am getting down voted for this because people haven’t tried it.
People here have tried it, LLM use is basically enforced at a ton of companies and the tech industry crowd is a huge portion of the population here.
You’re likely being downvoted as you seem to have a fundamental misunderstanding of how LLMs function. LLMs don’t reason. LLMs don’t think. Telling an LLM to not guess might help ever so slightly, but LLMs can and do hallucate search results regularly.
It’s a mistake to trust that LLMs aren’t hallucinating just because you told the LLM to not guess.
I juste don’t want to talk about this any more becausefor me adding those instructions in AGENTS.MD and giving it the full API and language and platform docs dropped significantly the hallucinations. You just disregarding my experience and want me to believe in your assumptions.
It helps, sure. But it still does stupid shit CONSTANTLY.
So do humans. Constantly.
It doing sgupid stuff is another issue separate from hallucinations. Preventing LLM from doing dumb shit is more difficult, you need more guards and harnesses.
This logically leads to a similar question to OP’s again: if it’s so easy to cut down on hallucinations, then why haven’t the designers of this software already integrated that in?
This is kinda fundamental to economics: the moment AI learns to perform some task, that task becomes less valuable.
But then aren’t there just different tasks to do? VisiCalc and Excel didn’t make accountants irrelevant.
That’s the economists angle, yes, but that’s not what I’m getting at.
You can’t make a business with AI because anyone can use an LLM to do anything an LLM can do.
There are exceptions in the short term, but they may not last. For example, we’re paying a hefty subscription fee for an AI service that scrapes financial data from scanned pages. It’s much more accurate than traditional OCR.
Presently its better than its competitors so commands a premium price. However, I don’t think it will enjoy that position for long.
AI is becoming a pretty big problem where executives are challenging experts or subject matter experts on topics based on their Ai usage. The problem is because the answers sound really good, but often the prompts aren’t quite right so then the experts need to do a bunch more work to explain and justify things that the executives just find rebuttals for through further ai prompting.
