Things That Make You Cringe.

Remove this Banner Ad

My friend got a message on a dating app that started with "Sure! Here's an opening line that sounds flirtatious without being overly familiar"
Did not smash
They've replaced the people in our training with a virtual AI trainer

So now i hear AI voice when watching training and they even sometimes show the AI trainer

Creepy almost person
 

Log in to remove this ad.

I must be the only person who has no interaction as far as i know with anything AI related. And never used chatgpt either

Sounds like im winning
I work for an American IT company that is all in on trying to make as much money out if AI as possible this year

They're also replacing as much of their own workforce with AI as they can

I've seen the money it costs and it's certainly not currently close to cheaper and i doubt it ever will be but hey everyone else is doing it
 
I must be the only person who has no interaction as far as i know with anything AI related. And never used chatgpt either

Sounds like im winning
Well now you know another person

I have nfi what they're talking about
 
I, too, had nothing to do with AI until a training program at work recently. Like anything, I see it as a useful tool if you use it in moderation. It’s certainly made a few laborious tasks, such as collating information or transcribing, easier but the moment you try to do anything creative with it, it’s just creepy.
 
I, too, had nothing to do with AI until a training program at work recently. Like anything, I see it as a useful tool if you use it in moderation. It’s certainly made a few laborious tasks, such as collating information or transcribing, easier but the moment you try to do anything creative with it, it’s just creepy.
There's 3 large problems with AI, well 4

There's no ethical way to create one, they're all built from stolen data

The massive energy cost of them, there's no such thing as moderation, they're the reason Google and Facebook and Microsoft etc have all walked back their emissions goals

Companies are trying to make them an everything solution, there are some tasks they are suited for but that's not what is happening

After a period of time they "hallucinate" meaning they become continuously less reliable and you can't train new models off AI data, it just collapses

So using AI for everything will break everything
 
There's 3 large problems with AI, well 4

There's no ethical way to create one, they're all built from stolen data

The massive energy cost of them, there's no such thing as moderation, they're the reason Google and Facebook and Microsoft etc have all walked back their emissions goals

Companies are trying to make them an everything solution, there are some tasks they are suited for but that's not what is happening

After a period of time they "hallucinate" meaning they become continuously less reliable and you can't train new models off AI data, it just collapses

So using AI for everything will break everything
Don’t disagree with any of that. Despite the training, I’m unlikely to use it for anything other than simple collating tasks. Energy costs is an interesting point, hadn’t considered that.
 
Don’t disagree with any of that. Despite the training, I’m unlikely to use it for anything other than simple collating tasks. Energy costs is an interesting point, hadn’t considered that.
The hardware they run this stuff on is massive racks of servers with enterprise GPUs with liquid cooling

The direct power consumption plus the heat they put out is massive

It's why the hardware vendors are all in, they make huge dollars selling the tin it runs on
 

Log in to remove this ad.

The massive energy cost of them, there's no such thing as moderation, they're the reason Google and Facebook and Microsoft etc have all walked back their emissions goals

The big tech companies have commissioned small nuclear reactors to power their data centres. They are claiming it's a reliable source of safe carbon-free energy.

 
The big tech companies have commissioned small nuclear reactors to power their data centres. They are claiming it's a reliable source of safe carbon-free energy.

There have been brown outs in towns where the big companies have AI farms in America
 
I must be the only person who has no interaction as far as i know with anything AI related. And never used chatgpt either

Sounds like im winning
Its useful in its place. Like many things, once its "the thing" it gets used well out of its place.
Maybe go to chatgpt and ask "what are things AI is really bad at?"



Come to think of it :

AI has made incredible strides in many areas, but there are still several things it's quite bad at. Here are some of the key challenges AI faces:


1. Understanding Context and Nuance


AI can struggle with understanding context in conversations, jokes, or sarcasm. It may misinterpret the underlying meaning of a statement, especially in situations where there's ambiguity, cultural references, or emotional undertones. For example, AI might not always grasp when someone is being sarcastic or when an expression is meant figuratively.


2. Common Sense Reasoning


Despite advances in machine learning, AI lacks true common sense. It often fails to make simple logical inferences that humans can do instinctively. For instance, while it might recognize that a person is sitting at a desk, it may not "understand" that the desk is for working or studying, or that someone might need space to write.


3. Creativity and Originality


While AI can generate creative content (like art, music, or stories), it is usually drawing on existing patterns and data. True creativity, the ability to create something entirely new or think outside established frameworks, is still a challenge. AI is often limited to remixing and combining things it's already been trained on.


4. Emotional Intelligence


AI still struggles to fully understand and respond to human emotions. While some systems are designed to recognize emotions based on voice tone, facial expressions, or text sentiment, they don't "feel" emotions or understand the complexities of human emotional experiences. This makes AI less effective in roles requiring empathy, such as therapy or deep personal interactions.


5. Generalization and Adaptability


AI often excels in narrow tasks but struggles with generalizing across different domains or adapting to new, unforeseen circumstances. For example, a model trained to identify cats in photos might not perform well when asked to identify different types of animals unless specifically retrained.


6. Ethical Decision-Making


AI has trouble with ethical reasoning, particularly in situations involving moral dilemmas or decisions with long-term consequences. Machines lack a moral compass and often follow rigid rules or data patterns without understanding the human implications of their actions.


7. Physical World Interaction


AI struggles with real-world tasks that require fine motor skills, flexibility, or the ability to adapt to unexpected changes. While robotics is making progress, AI in physical tasks (e.g., cooking, cleaning, or complex repairs) is still far behind human capabilities.


8. Bias and Fairness


AI systems often inherit biases present in their training data. This can result in unfair or biased decisions, particularly in areas like hiring, criminal justice, and lending. Despite efforts to mitigate bias, AI models still struggle with ensuring fairness and equality.


9. Long-term Planning and Strategy


AI can be good at optimizing short-term outcomes, but it often struggles with long-term strategic planning. Predicting complex future scenarios with multiple factors is beyond the current abilities of most AI systems, especially in dynamic, real-world environments.


10. Interpreting and Handling Ambiguity


Human language is full of ambiguity. AI can have difficulty when words or phrases have multiple meanings depending on context (e.g., "bank" as a financial institution vs. the side of a river). Without deeper contextual understanding, AI may not resolve these ambiguities properly.


While AI continues to improve, these areas represent some of the significant challenges it faces today. It highlights the importance of human involvement in areas requiring complex judgment, emotional understanding, and nuanced decision-making.
 
Reading about the Melbourne Victory's North Terrace efforts this weekend:


I gotta say the whole notion of football 'ultra' groups is bloody cringeworthy.
 
Finished work, made my way to Woolies.

Along the way I was doing some arm and leg stretch and flex movements. One such movement was imitating/ pretending to roll my arm over doing cricket bowling motions.

As I entered Woolies I did a half underarm fist pump while also motioning for the automatic doors to open.

I turn around proud of my efforts, where I see someone from my work likely followed me the entire way from point a to point b.

Nevertheless I am doubling down tomorrow and will be in tennis mode at work (both clothing and mannerisms).
 
Last edited:

Things That Make You Cringe.


Write your reply...
Back
Top