Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ed Zitron loses his mind annotating an AI doomer macro memo

dropbox.com

Post reply on HN
Print view View original

61–70 of 130 posts

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#61

Post by MadxX79 »

Ozzie_osman wrote: post #40
arowthway wrote: post #33

The financial market things are over my head and I don't have a dog in the game, but I think "Nobody is replacing salesforce with their internally vibe coded software" is just false? Both taken literally [0] [1] and as denying the general trend. Just in my company we already replaced WMS software subscription with own solution, and I wouldn't be able to write it fast enough and maintain it by myself without the use o…

Agreed. If anything, it puts downward pressure on pricing. Even if the CIO still buys Salesforce or whatever other tool, they won't be willing to pay as much.

If you don't give me a discount on my salesforce subscription I'll shoot myself in the face with this AI enabled gun?

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#62

Post by surgical_fire »

davorb wrote: post #31
Lariscus wrote: post #9

I do enjoy a good Ed Zitron sneer. The fact that the original article moved markets says a lot about the critical thinking skills of stock market traders.

You should look into how he destroyed the small indie MMO Darkfall and gave the game 2/10 without ever playing it, in a Eurogamer review a few years ago. The developers had receipts and could prove that he hadn't played it. It doesn't have any material effect on this article, but it says something about his ethics.

From Wikipedia:

> Darkfall lead developer Tasos Flambouras claims that game server logs show that the Eurogamer reviewer played the game for under three hours, a claim denied by the writer.

Even if we take the lead developer's word for it, what you are describing is simply false.

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#63

Post by MadxX79 »

simianwords wrote: post #34

Ed's main thesis is that cost is unsustainable for AI companies but this is clearly wrong. The unit cost is going down and has gone down by more than 20-30x over the years. Sure, the fixed cost of training is going up but that's because of the implied returns. Once the returns to training don't happen, it would simply reduce modulo cutoff date updates. The companies have a choice to just stop training and focus on in…

Is that the cost per token or the actual cost of the user having a conversation, reasoning and all?

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#64

Post by surgical_fire »

simianwords wrote: post #56
JanneVee wrote: post #52

Earlier quoted context omitted.

Well from my point of view. When they talk about gigawatt datacenters, then yes it is economically nonviable. You just need to know the scale of a gigawatt to realize that we need to start building power plants and fortifying the power grid to ship a gigawatt of power to a single location. Until the build out which takes years mind you, it is competing with other consumers of power. Lets take another huge consumer of…

but the costs of inference have been going down 20x to 30x over the years. so how can you tell it is nonviable? unless you are saying they are not paying market rate for the inference

I have no idea if costs indeed came down 20x-30x.

This claim sounds extremely fancy when AI companies bleed money, and will keep bleeding money in the foreseeable future.

I don't pretend to know the future. Maybe LLMs become economically viable and are the future, maybe not. I don't really care either way, to be frank.

And I use LLMs, btw. I pay for a ChatGPT account, but I find it only moderately useful. I always sort of question myself upon renewal date if it is worth the 20 bucks I spend monthly on it.

In no small part I keep using it to keep myself up to date on the best practices of using them in case it becomes standard.

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#65

Post by simianwords »

MadxX79 wrote: post #63
simianwords wrote: post #34

Ed's main thesis is that cost is unsustainable for AI companies but this is clearly wrong. The unit cost is going down and has gone down by more than 20-30x over the years. Sure, the fixed cost of training is going up but that's because of the implied returns. Once the returns to training don't happen, it would simply reduce modulo cutoff date updates. The companies have a choice to just stop training and focus on in…

Is that the cost per token or the actual cost of the user having a conversation, reasoning and all?

Cost per defined capability. Meaning you fix the task and then find how much it cost to achieve it including reasoning, tokens etc.

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#66

Post by simianwords »

surgical_fire wrote: post #64
simianwords wrote: post #56

Earlier quoted context omitted.

but the costs of inference have been going down 20x to 30x over the years. so how can you tell it is nonviable? unless you are saying they are not paying market rate for the inference

I have no idea if costs indeed came down 20x-30x. This claim sounds extremely fancy when AI companies bleed money, and will keep bleeding money in the foreseeable future. I don't pretend to know the future. Maybe LLMs become economically viable and are the future, maybe not. I don't really care either way, to be frank. And I use LLMs, btw. I pay for a ChatGPT account, but I find it only moderately useful. I always so…

https://epoch.ai/data-insights/llm-inference-price-trends

Do you have any reason to not believe it? It’s expected for costs to come down

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#67

Post by surgical_fire »

simianwords wrote: post #66
surgical_fire wrote: post #64

Earlier quoted context omitted.

I have no idea if costs indeed came down 20x-30x. This claim sounds extremely fancy when AI companies bleed money, and will keep bleeding money in the foreseeable future. I don't pretend to know the future. Maybe LLMs become economically viable and are the future, maybe not. I don't really care either way, to be frank. And I use LLMs, btw. I pay for a ChatGPT account, but I find it only moderately useful. I always so…

https://epoch.ai/data-insights/llm-inference-price-trends Do you have any reason to not believe it? It’s expected for costs to come down

The graph you linked seems to compare different OpenAI models in terms of "price per million tokens".

I am very skeptical of any financial information that comes from OpenAI. I have no idea how truthful those numbers are, or how creatively they can be collected to paint a rosier future for them.

Even if the numbers are truthful, I have no idea how the calculate price there. Is it in terms of cost of compute they rent? Is this cost subsidized or not?

Also, I don't know this "epoch.ai" website, I don't know their stance. The website name itself does not inspire my confidence on their reporting of anything related to AI. "Eat meat, says the butcher" vibes and all.

You can claim that the AI bleeds money because training is expensive, but inference is cheap. So it will only be financially viable when they stop training models? So they would need to stop improving their capabilities entirely for it to make any sense, is that your claim?

Even if I take this claim at face value (and that would take a lot of faith I don't have to give), it doesn't sound as good as you think it does.

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#68

Post by simianwords »

surgical_fire wrote: post #67
simianwords wrote: post #66

Earlier quoted context omitted.

https://epoch.ai/data-insights/llm-inference-price-trends Do you have any reason to not believe it? It’s expected for costs to come down

The graph you linked seems to compare different OpenAI models in terms of "price per million tokens". I am very skeptical of any financial information that comes from OpenAI. I have no idea how truthful those numbers are, or how creatively they can be collected to paint a rosier future for them. Even if the numbers are truthful, I have no idea how the calculate price there. Is it in terms of cost of compute they rent…

>To analyze the decline in LLM prices over time, we focused on the most cost-effective LLMs above a certain performance threshold at each point in time. To identify these models, we iterated through models sorted by release date. In each iteration, we added a model to the set of cheapest models if it had a lower price than all previous models that scored at or above the threshold.

Can you look at the analysis? It will make it clear. I mean its so obvious because GPT 4 costs way more than GPT 5.2-mini but much worse performance.

>Even if the numbers are truthful, I have no idea how the calculate price there. Is it in terms of cost of compute they rent? Is this cost subsidized or not?

Do you think they are subsidising 900x or simply that the costs have gone down?

Overall you have shown what I feel is extreme skepticism in something that is obvious. You can literally run a model in your laptop that matches an older closed model. Costs are obviously going down, I have shown data. Use your own anecdotes and report.

Extreme skepticism in such a way doesn't do any help.

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#69

Post by orwin »

simianwords wrote: post #44
zero_shift wrote: post #20

Earlier quoted context omitted.

Cost, debt, difficulty forming a moat, gap between what the product promises and what it can do, and the difficulty actually raising capital required. His style is acerbic and (imo) excessive sometimes. But he's also one of a minority of journos actually looking at the numbers and adding them up. Which seems to be a rarity

cost is going down 20x, 30x over the years so he's wrong about this.

That doesn't matter if the free models are as performant in 6 months. I will never personally pay for a model I can have for free. ChatGPT 5 used to be my preferred model as a DMing help tool, now deepseek and LeChat are the one I use, and are better at what OpenAI model use to be better at. And I think the models hit their limit for my usecase, I don't need better one. I never 'reprompt' anymore, and just roll/improvise with what I got.

Re: Ed Zitron loses his mind annotating an AI doomer macro memo

#70

Post by jubalfh »

hiddencost wrote: post #26

I've started to feel like Ed Zitron is actively hurting people I care about. I'm lucky to have worked in the field for a long time, and be able to spend a lot of tokens. In the last month it's become clear to me that the tech works. The science is done, and what's left is engineering. There are a lot of risks and mitigations and theory to build, but it's all solvable. The tech isn't mature, but neither was the Intern…

nice darvo, mate.
Post reply on HN