Is "AI will fix it" becoming the new default (and frustrating) answer for every business challenge?

Posted by Independent-Show-723@reddit | sysadmin | View on Reddit | 29 comments

I've been noticing a trend in various organizations, and I'm curious if you've seen it from the trenches: The default assumption that "AI will fix it" for virtually any business challenge, often before anyone has truly drilled down to the root cause.

It's like management hears "AI," and suddenly every structural inefficiency, communication breakdown, or outdated manual process gets rebranded as an AI opportunity. The risk isn't just wasted budget on complex solutions for simple problems; it's also diverting attention from critical systemic issues that a shiny new algorithm can never truly solve. You end up with an AI layer on top of a broken foundation, and guess who has to make it work?

I'm thinking about the value of having a go-to resource – essentially, a curated directory of proven, non-AI operational workflows and systemic solutions for common business problems. Something indexed by problem type, offering a baseline comparison. The idea is to help quickly identify if an issue is genuinely a structural inefficiency needing a process overhaul, rather than just a lack of automation.

Have you experienced this "AI-first" pressure? What are your thoughts on needing more accessible, peer-reviewed non-tech solutions to counter the hype and ensure we're solving the right problems effectively?

[-]

Kardinal@reddit

No. And it drives me nuts how many sys admins think that management thinks this way.

I'm senior enough to he in the room with the execs and the consultants about AI. That's not how they think about it.

[-]

Matazat@reddit

Wow yes you're very important and better than the rest of us. I work directly with SMB owners and they quite literally make business decisions based on things they saw on TV or read about on social media and they are chomping at the bit to try to get AI to do their jobs for them.

[-]

Last-Appointment6577@reddit

you've held that many jobs that you can safely say this?

[-]

Kardinal@reddit

I work across a number of organizations in my role. There's a tight network of companies that all do similar things and I'm at a level where I work with senior technical people from many of them. I hear what direction they're getting from their management.

I also am in the room with consultants who are advising or briefing our management about what to think about in terms of AI implementation. I hear the questions that are asked, and I hear what the consultants say their customers are concerned about.

What are they concerned about?

a) Accuracy. Is it good enough?

b) If it replaces too much of the staff, where do I get qualified staff for the positions it can't replace?

c) Is it really worth the money?

Sure, they're thinking about how it can save money. But they're not idiots. If nothing else, they know how much bullshit they sling around for their customers, of course they think about the way that their suppliers are bullshitting them.

AmoebaDue6638@reddit

The worst part is when the AI project fails, nobody says 'we should have just fixed the process.' They say 'we need better AI.'

i8noodles@reddit

my answer to "ai will fix it" would be "how?". if they cant tell me how then they dont actually know if it will. it will be entirely a gamble if it does.

Master-IT-All@reddit

The irony of a post asking about AI, from AI.

slop slop slop

catherder9000@reddit

What the F100 companies have been finding is that they can't find value, or even balance, from performance improvements or profit increases with the spending on AI and they're running scared. So, they've been laying off staff to show improved (or equalized) margins to keep the stock price from dropping.

The previous decade saw them laying off staff and using the wages for stock buybacks to increase the stock performance, and now it is laying off staff to pay for AI expenses to keep the stock from dropping.

Pretty soon it will just be management and marketing wanking each other off while nothing really gets done while customers migrate to some alternate solution because support, QA, and innovation have vanished.

North-Creative@reddit

I think that "soon" arrived quite a while ago. Even though there could be ai use cases, ai is mostly used to create more tech debt. Although I quite enjoy every companies notion of growth, which is to try and stay on the same level with ai and fewer people, when growth should mean they hire even more people due to the amazing ai results day and night.

thearctican@reddit

My company is even worse. We know exactly what our problems are and we’re being told from the top to “come up with AI ideas for the product”.

I don’t participate in the “invent-athons” because half of what is proposed can be done with single-network ML models or some genius says “we should have a chat bot”, and I’d rather not start a real argument in a big forum.

firesyde424@reddit

From the "I read it in a magazine" type of execs, yes.

cwk9@reddit

"Hey team, I read about this hot new company called Oracle. I need you to drop everything so we can do a POC next week".

Stop. This was literally how a conversation I had with my CTO started.

Generico300@reddit

Yes. Because when you are a non-technical moron whose only skill is bullshitting other morons, the ability to ask an AI to do something technical and then get a pseudo-functional solution from it feels like magic, and magic is the solution to everything.

But don't worry, those people will likely be out of business in a few years.

standish_@reddit

Peter's monkey paw curls

What if everyone is out of business, drowned in an ocean of shitslop?

largos7289@reddit

Just show them the news where AI decided that the whole car rental production database needed to be deleted. That should change their minds.

just read this, laughed the whole time.

Jazzlike-Vacation230@reddit

Management: I don’t know what to do, I don’t bother asking my employees cause I think they are idiots

Hey AI what do you think?

AI: Talk to your employees

Management: Lays off half the company in self inflicted confusion

chillzatl@reddit

Sure, over the course of a week our CIO went from looking for ways we could leverage AI to avoid adding staff to accepting that any AI based tools that could realistically do that require at least one resource to manage them that would cost as much as the people it replaced.

This trend is the same thing as it was in the early days of the internet, when the business world figured out that it was more than just a fad. It's also just like that "let's make an app" trend of a decade or so ago.

You combat it by using those as valid, real world examples of what jumping into technology trends because you feel obligated to without spending the time to really understand them can do to a business.

BrainWaveCC@reddit

Yes.

And there will be a market correction from this.

admalledd@reddit

We are nearing six months behind schedule on a corp-wide project because of the insistence on using AI to help plan/organize/do work.

One of the compatibility/risk reports had ~150+ bullet points of "risks/things to change for $Platform" and every single one was wrong because the AI instead found some similarly named thing on the public internet...

GinnyJr@reddit

People at our company are literally spending time trying to find reasons to “use AI” when they could just be doing their job

When is the next fad coming, because this one is getting really tiring..

Ouch!

Bubby_Mang@reddit

There's some tax incentives involved, though the sith council of management told me not to tell you guys that these kinds of breaks technically contribute to the bottom line, which would allow you to get a 5 on your annual review and thus be eligible for a raise. That credit obviously goes to the CFO.

TheVideogaming101@reddit

The COO at my company has unironically suggested replacing our SysEngineer team with "prompt engineers".

SASardonic@reddit

As somebody who deals in actual bespoke business process automation I feel like I would be morally obligated to punch somebody for saying this

Grant_Winner_Extra@reddit

AI won’t fix shit. People fix it and AI is a tool.

What you’re describing IS useful - if you’re consulting or developing AI.

N7Valor@reddit

AI is automation. Garbage in, garbage out.

If you have good ideas, that can be automated.

If you have bad ideas, that will also be automated.

orion3311@reddit

Eh, it does get old but what if you leverage AI to fix the broken foundation first? What if you leverage AI to build an automation that may not actually require AI in the process?

I agree on the buzzword burnout but some of these long-standing issues and tech debt we have might stem from just not having the time to fix them, but if we can leverage something to take hours or days into minutes or hours, thats a gamechanger.

All that said, you still can't fix stupid 😉