AI profitability

[JohnH]

In any case a sort in COBOL is done like this:

Code: Select all

SORT my-sort
       USING input-file
       GIVING output-file.

The premise of COBOL is that is a normal English sentence and easy for non-programmers to write and understand.

Yes, that's the premise. Which is why we have masses of COBOL code few people know how to maintain. What's hard about programming is not writing one line or one function.

Do you regularly program in COBOL? I haven't used it since perhaps 1982.

Maintenance has always been the hardest problem in software, followed by writing accurate specs in the first place. LLMs don’t directly address this, and actually result in great deals of software being written with no clear plan for maintenance.

Agentic AI does require writing clearer specs, so it’s an improvement from that side.

In any case, a sort in COBOL is in plain English just like writing an .md file for an AI.

[Bootstrap]

In any case, a sort in COBOL is in plain English just like writing an .md file for an AI.

That's true, I chose a poor example. There was already a very high-level abstraction for sorting. But the big deal was escaping assembly language. Writing a meaningful COBOL program is not a job for non-programmers. And the data structures were really a big step forward. But they are now at the heart of all programming languages.

And any substantial COBOL program is hard to read, hard to separate figure from ground, harder to maintain than the equivalent Python program. It IS a programming language, but it's hard to scan visually to quickly get an idea what is going on. All the verbosity means a lot more reading.

Just look at the job ads for COBOL programmers from those times. Or try this - convince Soloist he could write the same program in COBOL in a reasonable amount of time.

Here's where COBOL got us:

https://www.ciodive.com/news/IRS-old-so ... rt/642577/

I guess they couldn't find enough non-programmers to fix the problem?

[mike]

I think you guys are demonstrating another bit of opinion I heard today, which is the idea that AI is not going to take away all jobs, but it will fundamentally affect how many jobs are done. It won't change the job or the purpose of the job, but it will change the tasks that are performed. And it will bring about huge productivity gains for those jobs.

Arguing that this will destroy the economy would be like arguing at the time that the Industrial Revolution would destroy the economy. It will be disruptive, it will not be useful for every category of labor, it will present as yet unknown challenges for humanity, but it will not destroy the economy.

I've heard it said that AI is a greater development for humanity than the Industrial Revolution, and I suppose that could be true when you think about how it basically utilizes and synthesizes the known sum of human knowledge in history, including the core technologies used in our modern world to perform tasks and retrieve and package information and so forth, and even develop new capabilities.

What I wonder is when we all use AI tools instead of all the old software tools, what will happen to the underlying core knowledge that has made all this separate, independent software work all these years. Will people still be around that know how to code? What will the next generation of software engineers study?

[JohnH]

The main driver of technological growth and “the economy” is war, so I guess the question is how much AI can do for the military.

[Bootstrap]

I think you guys are demonstrating another bit of opinion I heard today, which is the idea that AI is not going to take away all jobs, but it will fundamentally affect how many jobs are done. It won't change the job or the purpose of the job, but it will change the tasks that are performed. And it will bring about huge productivity gains for those jobs.

Arguing that this will destroy the economy would be like arguing at the time that the Industrial Revolution would destroy the economy. It will be disruptive, it will not be useful for every category of labor, it will present as yet unknown challenges for humanity, but it will not destroy the economy.

Economists always say that, but a lot depends on a lot of things. Factory workers did not go out and become programmers when rust belt jobs went away. There was a huge shift toward jobs that required college education, then that became unaffordable, and it never was for everyone. We have not traditionally honored or payed skilled workers enough. So there are winners and losers, but there is always work for those who do it.

I've heard it said that AI is a greater development for humanity than the Industrial Revolution, and I suppose that could be true when you think about how it basically utilizes and synthesizes the known sum of human knowledge in history, including the core technologies used in our modern world to perform tasks and retrieve and package information and so forth, and even develop new capabilities.

If AI looks like the computer in Star Trek, good things will happen. If it looks like Dave in 2001 A Space Odyssey, not so good. The jury is still out on that. More when I have time.

What I wonder is when we all use AI tools instead of all the old software tools, what will happen to the underlying core knowledge that has made all this separate, independent software work all these years. Will people still be around that know how to code? What will the next generation of software engineers study?

Software engineering. Design. Understanding the problem you are trying to solve. Grokking large systems and making them maintainable. You need every software engineering discipline in the book and then some if you do complex work with AI.

[JohnL]

I gotta wonder if this podcaster found himself a sponsor in the AI industry and they paid him to say that AI users were seeing bigger profits. It just all sounds like a slick sales pitch to me that I’d hear from a car salesman. Only instead of chrome bumpers or great gas mileage the bait is “real profits”. That’s just how I’m seeing it.

Yesterday I heard a podcaster who has had a pessimistic view of the economy, partly because of fears of an AI bubble, start to change his tune a bit. In his view, for some time the market has been inflating an AI bubble similar to the dot-com bubble of the late 90s or the housing/credit bubble leading up to 2008.

With AI being a new technology, it has attracted huge amounts of venture capital, and the idea was that eventually the better technologies and companies would be sorted out, and there would be huge losses on the part of those who don't succeed, driving a downturn in the markets.

But what he said was that industries connected to AI are beginning to make real profits. According to what I found by poking around a bit, OpenAI (ChatGPT) is seeing the fastest growth, and is making some of the greatest investments, but is only on a path to profitability by 2029-2030.

However, Anthropic (Claude) is nearing break-even status, seeing extremely high growth, and should be profitable by 2027-2028. And Google, for which AI is only a part of its business, is of course hugely profitable. And these are just a few of the larger AI companies of course.

What all of this means is that the actual productivity of AI is likely gaining momentum, and that what we are seeing in the growth of AI in terms of stock value may actually reflect that rather than a bubble ready to burst.

Just thought it was interesting.

[Soloist]

what I'm making won't be making any money. Have I learned some about Python? sure but to build a product to sell requires having an item to sell that has value to others worth paying for.