this post was submitted on 10 Apr 2024
73 points (97.4% liked)

Futurology

1784 readers
139 users here now

founded 1 year ago
MODERATORS
all 7 comments
sorted by: hot top controversial new old
[–] [email protected] 22 points 7 months ago (2 children)

As someone who loves ai and thinks it's an integral part of a new and better world for all -- Jesus christ what are these companies and journalists on?

Of course you can't just throw current ai into any old shit and magically save money, especially when they're talking about super basic LLMs that we have now. I don't even know what the author is trying to say tbh, a lot of wikilinks to buzzword organizational issues and a false assertions about what ai is?

Yes giving everyone in management a hammer is a waste of money, yes giving the shop guys a hammer is useful.

[–] [email protected] 12 points 7 months ago

That's how management works, though. It's never about doing something long term beneficial, but always just seemingly good things for the next quarter. Using AI, or even just announcing to use it, boosts the stock price and the career of whoever is in charge. That it's all crashing down a few months later isn't announced anywhere and swept under the carpet.

[–] [email protected] 2 points 7 months ago

It gets click from the redditors who hate anything AI.

[–] [email protected] 5 points 7 months ago (1 children)

An interesting read, listing several faults of organizations, and a conclusion that is really a surprise to no one.

In short, it would be wrong to think that a tool as rational as AI, in an environment as irrational as an organization, will automatically increase efficiency the way managers hope it will. Above all, before thinking about integrating AI, managers need to ensure that their organization is not stupid (in terms of both processes and behavior).

[–] [email protected] 4 points 7 months ago (1 children)

Both are irrational, though.

[–] [email protected] 2 points 7 months ago

Integrating machine learning (which itself is STILL a stretch even for LLMs) is absolutely a smart idea ... when used appropriately for dumb tasks that don't really matter if they screw up, or are otherwise under direction and control of a person, they make a lot of potential sense.

... attempting to replace actual employees with dumb tools marketed as intelligence ... is fucking moronic.