The A.I. Clusterfuck: Difference between revisions

From ms-demeanor.com
Jump to navigation Jump to search
No edit summary
No edit summary
 
(3 intermediate revisions by the same user not shown)
Line 1: Line 1:
I am not unreservedly pro-AI, however I have a lot of problems with how AI is criticized and reported on. I think there is huge potential for AI to exacerbate a number of problems but I believe that the root of the problems that most people are concerned about with AI are actually problems caused by capitalism.
I am not unreservedly pro-AI. I'm not even reservedly pro-AI. I am deeply AI-skeptical. In fact I think that AI is the mid 2020s version of "web 3" in terms of hype and over-investment. In spite of that, I have a lot of problems with how AI is criticized and reported on by people who are anti-AI. I think there is huge potential for AI to exacerbate a number of problems but I believe that the root of the problems that most people are concerned about with AI are actually problems caused by capitalism.
 
The TL;DR of this whole page is "there's a lot of nuance in this discussion that a lot of people are talking past, I suspect that AI is neither as helpful nor as catastrophic as it has been made out to be, and regardless of all of that I don't like AI very much and wish people used AI less.


=Defining Terms=
=Defining Terms=


=Concerns=
How you feel about AI usually has a lot to do with how you define AI. People who dislike AI the most generally tend to center their discussions around text- and image-generation AI.
 
It is difficult to draw a bright line in the sand between "highly tuned, specifically trained scientific model that predicts tsunamis via machine learning" and "gimmicky novelty slop image machine trained on copyrighted materials with a .ai address." It would be nice if there was a very noticeable dividing line, but there is not. Machine learning (ML) principles are used in creating generative AI tools, generative AI tools that are widely adopted contribute to advances in ML research. You can't grow your knowledge of healing without also growing your knowledge of suffering.
 
For the purposes of this conversation, the "AI" under discussion is largely going to be:
 
*Gen AI tools that produce text and images
*AI chatbots and web search tools that are either standalone or integrated into other tools
*Tools that are available to consumers on a broad basis
 
Any discussion of scientific research tools or ML models that are not available to a broad group of users outside of institutional or research access will be noted.
 
 
=General Concerns about AI=
At the moment (late July 2026)
 
==Security==
==Security==
===Myth, Fable, and AI Superhacking===
===Trusting AI to Keep you Safe===
==Environment==
==Environment==
===Water Use===
===Power Use===
==Datacenters==
==Datacenters==
===Datacenter vs Hyperscaler===
===Datacenter builds over time===
==Economics==
==Economics==
===Over-Investment and being "Too Big to Fail"===
===Component Hoarding===
==IP and Copyright==
==IP and Copyright==
===Fair Use===
===Fair Use===
===Data Scraping===
===Data Scraping===
==Data Control/Data Leaking==
==Data Control/Data Leaking==
===People Making Bad Decisions===
===AI Companies Being Genuinely Bad===
==The Military Industrial Complex==
==The Military Industrial Complex==
===Captcha used you to train robots how to bomb children===
==Cognitive Concerns==
==Cognitive Concerns==
===Does AI make People more Stupid?===
===How does AI impact Critical Thinking?"
=="Real" Art==
===Soul===
===Effort===
==AI as a Labor Issue==
==AI as a Labor Issue==
===Can an AI do your job?===
===Does your boss THINK an AI can do your job?===
===Will an AI take your job even if it can't do your job and your boss doesn't think AI can do your job?===
===Will AI make your job unsupportably miserable?===
==Slop==
==Slop==
=My Specific Concerns about AI=
==Economics and Tech==
==Cognitive Concerns==
==Art==
=Criticizing Media Criticism of AI=

Latest revision as of 13:42, 27 July 2026

I am not unreservedly pro-AI. I'm not even reservedly pro-AI. I am deeply AI-skeptical. In fact I think that AI is the mid 2020s version of "web 3" in terms of hype and over-investment. In spite of that, I have a lot of problems with how AI is criticized and reported on by people who are anti-AI. I think there is huge potential for AI to exacerbate a number of problems but I believe that the root of the problems that most people are concerned about with AI are actually problems caused by capitalism.

The TL;DR of this whole page is "there's a lot of nuance in this discussion that a lot of people are talking past, I suspect that AI is neither as helpful nor as catastrophic as it has been made out to be, and regardless of all of that I don't like AI very much and wish people used AI less.

Defining Terms

How you feel about AI usually has a lot to do with how you define AI. People who dislike AI the most generally tend to center their discussions around text- and image-generation AI.

It is difficult to draw a bright line in the sand between "highly tuned, specifically trained scientific model that predicts tsunamis via machine learning" and "gimmicky novelty slop image machine trained on copyrighted materials with a .ai address." It would be nice if there was a very noticeable dividing line, but there is not. Machine learning (ML) principles are used in creating generative AI tools, generative AI tools that are widely adopted contribute to advances in ML research. You can't grow your knowledge of healing without also growing your knowledge of suffering.

For the purposes of this conversation, the "AI" under discussion is largely going to be:

  • Gen AI tools that produce text and images
  • AI chatbots and web search tools that are either standalone or integrated into other tools
  • Tools that are available to consumers on a broad basis

Any discussion of scientific research tools or ML models that are not available to a broad group of users outside of institutional or research access will be noted.


General Concerns about AI

At the moment (late July 2026)

Security

Myth, Fable, and AI Superhacking

Trusting AI to Keep you Safe

Environment

Water Use

Power Use

Datacenters

Datacenter vs Hyperscaler

Datacenter builds over time

Economics

Over-Investment and being "Too Big to Fail"

Component Hoarding

IP and Copyright

Fair Use

Data Scraping

Data Control/Data Leaking

People Making Bad Decisions

AI Companies Being Genuinely Bad

The Military Industrial Complex

Captcha used you to train robots how to bomb children

Cognitive Concerns

Does AI make People more Stupid?

===How does AI impact Critical Thinking?"

"Real" Art

Soul

Effort

AI as a Labor Issue

Can an AI do your job?

Does your boss THINK an AI can do your job?

Will an AI take your job even if it can't do your job and your boss doesn't think AI can do your job?

Will AI make your job unsupportably miserable?

Slop

My Specific Concerns about AI

Economics and Tech

Cognitive Concerns

Art

Criticizing Media Criticism of AI