I strongly believe AI will drive efficiency. I’m less certain about whether that’s already happening or will happen in the future and when in the future.

However, I’ve already seen a CTO of a F100 company explicitly state that whether AI is driving efficiency or not, the capital investment, and more importantly, the promises of efficiency to investors will mean some people will be let go.

Efficiency is output/input. The input is easy to measure. It’s cost and in this particular case salaries.

Output is a lot harder to measure, which means it can be fudged easily.

So you cut the easily measurable inputs and inflate the easily manipulable output.

One could imagine the reverse would also be possible, where you maintain inputs but inflate the output, but there is an asymmetry where investors will reward you for cutting costs even if there are no efficiency advantages that makes cutting inputs more sellable than inflating outputs.

> Output is a lot harder to measure, which means it can be fudged easily.

First and foremost, this is about Oracle. For the short period I worked there, my impression about culture and tech was: mediocre. Not excellent, not poor but just a around average.

Which raises the question: why is it such a successful company commercially? I believe it's being ruthless to customers, employees and suppliers combined with cooking the financials.

Which bring me to your remark about output being difficult to measure. Imho Oracle had been exceptionally good at manipulating and obfuscating their output. And this was true long before AI came to the scene.

Oracle is mediocre because it could be.

That's probably where you'll end up if you're a company where that's an option.

No one was ever going to switch databases to a better DB as long as:

1) they did the bare minimum to ensure no alternative existed where switching made any sense.

2) they never charged too much where it made sense to switch thinking along the lines these businesses made decisions.

It's like the resource curse played out on a company scale.

If you have no incentive to get better, you won't.

> Which raises the question: why is it such a successful company commercially?

Tons of mediocre enterprise software is built on Oracle DB.

Why enterprises are vendor locked-in? There are very few large enterprise players in every industry that implement all kind of ISO, standards, got an army of business analysts to generate million of requirement pages. Any new player must fight against that artificially overblown legacy systems, design and prove migration process is possible etc.

Here are just a bunch of industries my family/friends worked in and had first hand experience with these legacy systems - Airline (PSS), Banking, Healthcare, Hotel, Telecom.

Sad, but true.

And if they had it their way, Oracle would have similarly strangled every last customer of Java, MySQL, OpenOffice, Solaris, etc. to squeeze out every last dollar.

And then make it look like they're innovating.

> a CTO of a F100 company explicitly state that whether AI is driving efficiency or not, the capital investment, and more importantly, the promises of efficiency to investors will mean some people will be let go

That seems like an insane gamble to me. Lay off all the workers now and hope that AI can deliver on its promise to replace them some time in the indeterminate future.

I think we wont know until the true costs for ai are revealed. Right now were still in the vc growth above all things part of the cost curve. It will get worse quality as revenue demands increase (as all products suffer).

It will be another dependency for all companies to bear. Hopefully significant gains for humanity, tbd

… and in the background the next hiring wave slowly builds into a tsunami.

Sometimes I feel like the only person left who remembers the pre-dotcom vibe. OpenClaw etc. should have set off alarms that we are back in the land of the Quick and the Dead.

Yeah OpenClaw for me was the alarm sounding... not just because of the project itself, but the fight between Altman and Zuck to pursue him. It shows a fundamental lack of product sensibility/visionary thinking within OAI and Meta. Having lots of dosh clearly doesn't solve that problem in-house.

You don't even have to inflate the output. You just cut the dead weight, which reduces input without harming the output.

In Oracle's case this is part of their larger Oracle Cloud strategy - once you remove the scooby doo monster mask you end up finding much of Oracle's AI spend was connected with their larger bet on Oracle Cloud and becoming a tier 1 hyperscaler.

This is why they landed one of the mega cybersecurity companies (the one who's name starts with C) as well as a globally distributed ridesharing businesses with sweetheart terms.

Oracle Cloud already represents 50% of Oracle's total revenue, and the AI story helps them justify that capex needed to fund their pivot into becoming a hyperscaler.

Everyone is holding their breath praying AGI either will, or will not, come before the chickens come home to roost.

Not sure that makes sense. As amazing a technical feat as AGI will be, does it follow that s.tons of money will immediately be made? That's not really how humans act, historically. Any migration to new technology takes years, decades. There are still steam engines pulling revenue service trains.

Yes - if AGI is made, nearly immediately every knowledge worker’s labor value goes to zero over night.

This may or may not include AI researchers.

That seems like a wrong economic theory to me. The economy is based on differential of value. I can make furniture, so it's worth less to me than it is to you. Therefore we can trade. That's what supply/demand is. If AGI somehow exists, then the value of intelligence drops to zero for both of us, there's nothing to trade.

AGI would not make knowledge work valueless, it would move all the knowledge work value to the AGI companies.

> AGI would not make knowledge work valueless, it would move all the knowledge work value to the AGI companies.

Yep! That's the point :)

There still plenty to trade - you just don’t have any of it.

Those $300 billion dollar circular deals will become much more common.

Can you guys just say the quiet part out loud?

That you want slaves. You want slaves. This is what you are asking for.

Unless you're paying the AGI? Then why not just... pay a human that is already present? Much more efficient.

Not sure if you’re insinuating that I’m one of the capitalists? I’m not, that’s why I’m saying it out loud. I believe capitalism’s relationship with labor will need to fundamentally change if AGI and robotics takeover. I don’t know what that looks like but obviously the current capitalist overlords want everything to stay the same - just for them to have more power/money.

You are assuming that the AGI service is very cheap..

Based on current trends why wouldn’t it be cheap?

Why would an artificial intelligence want to do what you tell it to?

I don’t think AI in its current form and current track wants much of anything.

Why would it have any desire at all?

And probably not much longer after that, the bloodshed

Unless the benefits of AI get communalized (UBI, or some other form of sharing), ya.

How would that work? Taxes are already full of perverse incentives. Certainly anything resembling UBI will be just as bad.

You’re not wrong - I was just giving that as an example but not saying it’s the right thing to do.

I don't think there is any compelling reason to believe this will actually happen

It will drive capital into data centers, which is good for Oracle.

Or maybe it won't. If it can be made efficient like humans it might be at the edge mainly.

What is this "everyone" running from?

Collapse of return on investment

don’t forget that they will probably hire thousands off-shore

Maybe - but the larger issue is FCF - and you can't hire that many without it showing up on FCF.

why would they try to hide it? no other company does :)

You must not understand what FCF is. If you're cutting headcount STRICTLY to juice FCF, you cannot spend money offshore without hitting FCF.

you can if you are paying pennies on the dollar :)

Except we're not, in this case. Oracle is not hiding the fact that the cuts are due to rising costs and expenses, rather than actualized efficiency gains. This is in contrast to everyone else, ex. Dorsey's block, who are putting a brave face on cuts, saying it's simply due to efficiency gains from AI. The question is: why are their statements being taken at face value?