Diminishing Returns in AI: The Most Common AI Mistake

Written by

admin

in

, , , , , , , , , , , ,

At some point, more of the same stops paying off … it is called the law of diminishing returns.

Law of Diminishing Returnsvia Sketchplanations

Nature (and common sense) reminds us that equilibrium is important. For example, when you exercise too much, you get injured; when you drink too much water, you get poisoned; etc. 

This concept applies almost everywhere.

  • It's why diversification is so important in portfolio construction theory. 
  • Or, why you don't want to put all your eggs in one basket (concentrating your risk).
  • And, my favorite, it's also why you shouldn't only eat vegetables.

A related nugget of wisdom from the extreme … Too much of a good thing is a bad thing! 

And of course … Be moderate in everything, including moderation.

A recent study on the effects of ChatGPT use on brain activity also supports this theme. 

via "Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task" 

The study divided 54 subjects into three groups, and asked them to write several SAT essays using ChatGPT, Google, or nothing at all. They used an EEG to record the writers' brain activity. ChatGPT users showed the lowest brain engagement. Over time, ChatGPT users got lazier. 
As a result, the ChatGPT essays also ended up lacking original thought and sounding similar. 
 
The paper suggests that the use of LLMs could harm learning and retention. The group that wrote without any help showed the highest neural connectivity, and higher brain activation in creativity ideation, memory load, and semantic processing. This group was more engaged and curious, and expressed higher satisfaction with their work. 
 
The paper hasn't been peer-reviewed yet and has a small sample size. 
Ultimately, I think it's telling us something we probably could have already guessed. If you're not actually the one reading or writing something … you're not going to absorb the information as well. Your brain isn't going to activate as well. And if you're not finding other places to train your brain, you'll get worse at critical thinking. 
 
A few weeks ago, I wrote about my research process and how AI has both bolstered and hindered certain aspects of it. I'm casting a wider net, but going less deep. 
So now, I’m exploring how to maintain a balance. I still want to leverage AI’s value, while reintroducing a layer of slowness and reflection into the process. Maybe that means manually summarizing some articles. Maybe it means pausing to journal about what I’ve read, or discussing it with someone. The goal is not to abandon the efficiency — but to ensure that efficiency doesn’t come at the cost of depth. - via "Is AI Making Your And Your Team Smarter?"
This isn't a call to avoid AI or to slow down its development. It's a call to deliberation and intentionality. Students are flaunting that they're using ChatGPT to shirk the system. Companies are releasing AI-written copy not only on their websites but also in press releases. Humans are falling in love with chatbots and leaving their spouses. 
 
Meanwhile, AI is also the underpinning of countless enhancements and increased productivity. 
 
The same thing happened with the Internet. 
 
At a certain point, overreliance becomes a crutch, not a tool. 
 
I bet AI makes unintelligent and lazy people lazier and less intelligent. 
I'm positive it can make YOU smarter and more effective if you use it in moderation and to support your unique abilities. 
Onwards! 

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

More posts