Are Large Language Models Really AI?

They pay these people almost nothing, and they still want to replace them.

To become a Billionaire in the first place you are going to be far more likely to have negative psychological traits. That's how they get to where they get to.


These are the people we are hoping will be nice to us, when we collectively no longer have any leverage over them. Not saying it's gonna happen, but it's interesting to think about.

Most tech companies (or even game companies) are not proper companies, they do not function based off profits. They function based on funding rounds (Discord, Reddit, Open AI etc), they effectively just keep begging for money from mega wealthy people and using snakeoil salesman tactics to get money to fund the company well beyond their means. Many of these companies make zero profits.

Open AI did vast funding rounds and raised huge amounts of money, and with that money they agreed to buy up 40% of the worlds ram to lock out competitors for instance. A corporate espionage tactic, funded by money they could not make themselves (they are currently imploding).

This is why they were hiring so many people 5+ years ago, and why these companies are full of people literally doing nothing (you see endless women on tiktok working at tech companies in manufactured jobs, making videos about doing nothing and eating avocado and drinking starbucks). Cause the inflated work force implied growth and people invested cause of the implied growth from the bloated workforce (this is also why they have so much DEI of incompetent people, cause they don't do anything to matter and it helped gain investor money).

The are now downsizing with AI for the same reason, to pander to investors that AI is the future that they are embracing and that will improve the value of the company long-term.

However, ones making profits and not begging for money are actually doing it for real reasons.

You are confusing tech startups with established tech corporations, those are wildly different things.

Tech startups are entities working towards building something of value and while profits are very nice they really just want value as their ultimate goal is to sell that value to someone else. That someone else is either a larger tech company like Microsoft, Broadcom, Meta or Amazon, or to the public via IPO. Established tech companies are those that have already transitioned into a full profit model by leveraging whatever products or offerings they have.

This is discord, because you mentioned them.

https://www.businessofapps.com/data/discord-statistics/

Total investment is $1bn USD, total valuation is $14bn USD, revenue is around $500~600 million USD, though they are still operating at a loss. Back in 2024 they terminated a large number of staff to reduce operating expenses and I suspect they are going to start selling gaming usage data as a secondary revenue stream in preparation for either an IPO or a sale.

Ever noticed how most of us have discord starting automatically and it displays which game you are playing? Yeah they are in a key position to sell gaming usage data either directly to game publishers or to data brokers who will then sell it to game publishers.

OpenAI wasn't even a for-profit company. It was founded and funded as a non-profit company then back in 2019 pull a fast one and "restructured" by starting a new for-profit company and transferring all the IP into that company. Right now they are a privately held company and I'm pretty sure they are being used as a debt bag for lots of this AI circular investing nonsense.

After studying a little more about vibe coding, I conceded.

It does make sense having a scrum master agent…

Before the great recession mass layoffs usually increased stock prices. This may seem odd but the stock market reflects expectations and emotions rather than anything like reality or P/E ratios.

I think lean and efficient has replaced lean and mean. Both are reductions in personnel expenses with a positive coat of paint.

As someone who has often made my living sewing visuals of people sewing at normal sewing machines is NOT going to train GP robots to sew. And it certainly is not going to teach cutters to cut.

We are closing in on fully automated sewing. I mean fabric in, clothing out. There are no machines that look like anything I have ever sewn on and no more GP robots that on an automotive production line.

How Automation In Apparel Manufacturing Is More Than Humanoids: The Tech Innovation That Matters
Forbes

Automated Cutting & Sewing For Industry 4.0
Textile world, 2/2024

That's my point, prior to the GFC companies could do mass layoffs with the expectation that "line go up". But nearly two decades later account managers are a lot more skeptical and it's all about the wording on "why". Publicly traded companies can't lie on those notices, but they can be economical with the truth, which they usually are now.

I really like that "economical with the truth" line. Its much more elegant than mislead.

Yeah SEC regulations make it illegal to outright lie on public mandatory public disclosures. So instead they overemphasize things that make them look good while underemphasizing things that don't. Like claiming their "AI initiatives" are expected to bring this massive productivity boost and therefor they don't need as many people.

That statement itself is not a lie, they executives can "expect" that, and expectations are wrong all the time.

This scene from Wolf of Wall Street summarizes the whole thing.



Nothing said in that phone call was a lie, it was all complete truth, and also highly suspect to anyone not keen enough to catch the words.

More AI nonsense.

Shares in Allbirds surge after maker of wool sneakers announces pivot to AI

Rebrand as NewBird AI sent shares up 582% in bizarre and rapid turnaround for firm that had fallen on hard times
What role can AI have in the production or sale of overpriced sneakers made from merino wool?

I went into a rabbit hole yesterday.

Now I really don't like Larry Elison, Peter Tiel and Palantir.

You don't need a rabbit hole for that.

Now your starting to get into the real life versions of the lizard people.

I love "vibe coders". Love the vibes from all the fat contracts to delete their vibes and replace it with non-vibing code that actually works without setting things on fire.

Billionaires in the AI age be like

"dear sir or madam, please justify your existence!"

It does where I went with it.

Around the Apple campus, this is known as being Steve Jobs-ed

https://www.forbes.com/sites/lanceeliot/2025/09/18/the-new-job-of-being-a-vibe-coding-cleanup-specialist-is-intriguing-and-stirring-ample-controversy/

https://www.linkedin.com/posts/dani-markovits_plot-twist-ai-was-supposed-to-kill-developers-activity-7373285928515149824-11wT/

Whole new career field called "Vibe Coding Cleanup Specialist", which is just an "AI enabled" way to say Senior Development Engineer.

MOAR ELECTRICITY FOR MY DATA CENTERS!!

Somehow I can believe that.

Companies love virtue signalling. Especially virtue signalling they can get employees to do on their own time on behalf of the company without paying them.

A certain extremely large multi national bank sent around an email to all tech staff that a "priority" of the year was to change default branch name to "main" from "master" for $reasons.

Nothing better to do with employees' time than virtue signal to the woke masses. Nothing more important than virtue signalling, to the point that the CTO specifically mentions it as priority for every team to do in the year.

Result - none. Just virtue signalling.

Opposing it, privately or publically, does you no good.

The campaign contributing billionaire class wants low wages but they don't care how they get them lower. If they could buy slaves they would but they can't therefore the push to robotics.

Many American politicians worry loudly and publicly about the low birth rate. A number of them blame it on the lack of teen aged pregnancies. Other than saying those ones should be shot out of cannons I can't go further into the subject because politics and off topic to boot.

It is, however, off topic unless you have details on how its going to be run by AI.

P. S. Please have those details.

Talking about AI, interesting how many youtubers are being demonetized by blurry allegations.

You know? I cant help but feel like Youtube is walking a path very similar to FFXI.

Like, when they find something to exploit, the idiots leak the exploit and the others starts abusing it like no tomorrow, forcing SE to take measures to annul or mitigate the damage.

AI was the "exploit" to Youtube. Impressive the amount of the so called "dark channels" become a trend. Many and many videos on Youtube teaching how to automate the entire video production with AI. Those guys could go from picking an obscure niche, generate an AI script, AI imagery, AI voice, wallmart edition, AI thumbnail and AI title, pack it all together and release dozens of videos per week.

And all of them getting lots of views.

The official allegation is that using AI is generating "inauthentic" content, what is true, given you have a blueprint to how to get those videos e views. But I suppose having so many shady people literally farming real money out of Youtube must be hurting Googles pocket more than saying something is not authentic.

Meh, in parts I kinda agree with Youtube, mass producing videos using cheat codes does feel cheap. But they are also netting views so it is not entirely wrong, unless they are also automating the views.

I think in the end, my stance is that videos, music, imagery, are all forms of communication.

And communication are only meaningful if done between humans.

I can accept a fully AI content, if at least the script has a human meaning and was writing by a human.

Now, a message fully written by machine, this I feel garbage.

Like my work, when they play some AI generated song.

On the other hand, the images of the AI generated muscular dog has a meaning, it portraits my identity. Yes, its an AI image, but you know who Im just looking at the image, so this AI content has a meaning

I asked Claude to write shitty code for me today so I can demonstrate to others what shitty code looks like.

It did a great job.

How can you tell it's written shitty code as opposed to its normal code output.