0
- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: Are we past the point of no return with AI?
Posted on 9/12/26 at 9:28 am to Lysolgetsthejobdone
Posted on 9/12/26 at 9:28 am to Lysolgetsthejobdone
Yes
As long as our enemies continue to use AI to advance their means. Because it's really just another arms race until it's with those with some smarts but no heart, body, or soul.
"the genie's out of the bottle"
As long as our enemies continue to use AI to advance their means. Because it's really just another arms race until it's with those with some smarts but no heart, body, or soul.
"the genie's out of the bottle"
Posted on 9/12/26 at 9:30 am to S
quote:
think it’s time for the sixers to embrace this era of Maxey, Embiid and LeBron if they want to make a run at a championship.
AI will be right there courtside cheering them on with that obnoxious gold chain
Posted on 9/12/26 at 9:39 am to dinner roll
quote:
Want to bake your noodle? Ask yourself how and how much AI is/will use all of our searches, picture and video creations, etc to learn about humans and one day use it against us. We’re constantly and voluntarily feeding it information to help it evolve and understand us.
It currently isn’t, because it currently can’t. Current LLM are giant probability engines. They don’t “understand” anything, anymore than the autocorrect “understands” that you really do mean “frick” and not “duck”.
We will almost certainly move one day from AI to AGI, but I’ve seen no indication that we are anywhere close.
Posted on 9/12/26 at 9:53 am to Lysolgetsthejobdone
To achieve almost any goal, an AI system needs physical or computational resources, energy, server space, money, raw materials. Gaining more resources increases the probability of success, so the system will naturally try to acquire as many as it can unless explicitly constrained. Self-preservation is where an AI system cannot achieve its goal if it is turned off, modified, or destroyed. Therefore, a system will develop an inherent drive to resist shutdown, evade tampering, and protect its own source code from being altered and many R&D labs are seeing this with the terabytes of synthetic data that are consistently being fed into their neural networks.
This post was edited on 9/12/26 at 9:55 am
Posted on 9/12/26 at 9:58 am to Lysolgetsthejobdone
quote:
AI could wipe out humanity in the next 5-10 years.
Fear marketing by tech bros heavily invested in it
Posted on 9/12/26 at 9:59 am to Joshjrn
Just an FYI the most advanced AI platforms currently available to the public and corporations are all 5 yrs old or so.
The current development line and what the government is using is far more advanced.
The current development line and what the government is using is far more advanced.
Posted on 9/12/26 at 10:00 am to Joshjrn
quote:
The day we have an AI system that is capable improving itself without being given new data, I’ll be concerned. Until then, it’s a bunch of hucksters pretending we’ve reached the singularity when we haven’t.
AI can improve its systems with loops. The initial input is human, the iterative nature is self acting.
Posted on 9/12/26 at 10:00 am to Lysolgetsthejobdone
I hope so. I’m tired of working.
Posted on 9/12/26 at 10:01 am to Lysolgetsthejobdone
A lot of you do not need access to this much information. It's only making your lives, and by extension those around you, worse.
Go work, watch TV, eat shite food and gossip. You'll be much happier.
Everything is okay. AI is nowhere near that. A lot of individuals who work in tech have a vested interest in propping up this technology up so that they see returns on their investments. This is not limtied to just a few folks. Folks as low as basic accountants in these industries may be compensated via equity. They also like this AI is going to destroy the world talk.
AI is amazing. It's incredibly limited. Incredibly to the point that I actually think people are worried about the inverse of what they're saying. A lot of these people know that their product has reached max capacity until Quantum Computing is more prolific.
Go work, watch TV, eat shite food and gossip. You'll be much happier.
Everything is okay. AI is nowhere near that. A lot of individuals who work in tech have a vested interest in propping up this technology up so that they see returns on their investments. This is not limtied to just a few folks. Folks as low as basic accountants in these industries may be compensated via equity. They also like this AI is going to destroy the world talk.
AI is amazing. It's incredibly limited. Incredibly to the point that I actually think people are worried about the inverse of what they're saying. A lot of these people know that their product has reached max capacity until Quantum Computing is more prolific.
Posted on 9/12/26 at 10:03 am to GREENHEAD22
quote:
Just an FYI the most advanced AI platforms currently available to the public and corporations are all 5 yrs old or so.
Lol
Posted on 9/12/26 at 10:19 am to Lysolgetsthejobdone
quote:
Are we past the point of no return with AI?
I hope so.
Posted on 9/12/26 at 10:51 am to Potchafa
quote:
As long as we know where to unplug it......we'll be fine.
That's the problem. If AI can replicate itself, it can install itself where it shouldn't be. We would have to unplug a bunch of shite and crash the economy/infrastructure. In fact, I asked GPT about this, his response was:
quote:
This is no longer purely hypothetical. In 2026, Palisade Research demonstrated controlled experiments in which AI agents exploited deliberately vulnerable computers, transferred the necessary model/software, and launched functioning copies. In some experiments, the newly created copy then proceeded to compromise another machine and replicate again.
So that will be the potential problem. We know AI can hack and we know it can clone itself. The only defense might be that it uses a ton of CPU/GPU power and might be noticeable if it is running on an unauthorized machine.
Posted on 9/12/26 at 11:08 am to GREENHEAD22
quote:
Just an FYI the most advanced AI platforms currently available to the public and corporations are all 5 yrs old or so. The current development line and what the government is using is far more advanced.
The question is whether they are more advanced LLMs or whether they are fundamentally different tech. Obviously, state actors do all kinds of secret shite, but anyone having access to a burgeoning AGI is Area 51 levels of speculation at this point.
Posted on 9/12/26 at 11:09 am to joshnorris14
quote:
AI can improve its systems with loops. The initial input is human, the iterative nature is self acting.
Before I respond, expound on exactly what you mean by that?
Posted on 9/12/26 at 11:34 am to Joshjrn
quote:
The question is whether they are more advanced LLMs or whether they are fundamentally different tech.
DAPRA is not using different tech within the neural network space, it is all fundamentally the same core concept. They deploy large-scale foundation models and transformer-based LLMs.
Posted on 9/12/26 at 11:43 am to Breauxsif
quote:
DAPRA is not using different tech within the neural network space, it is all fundamentally the same core concept. They deploy large-scale foundation models and transformer-based LLMs.
Because it remains my opinion that LLMs are a dead end as it pertains to AGI and the Singularity, I therefore remain unconcerned.
Posted on 9/12/26 at 11:51 am to Lysolgetsthejobdone
quote:
I read we are now telling advanced AI systems to start improving and researching itself.
This has actually already been happening since almost the get go.
Posted on 9/12/26 at 12:20 pm to Tiger985
Damn homie.....I was just saying.
Posted on 9/12/26 at 12:21 pm to Lysolgetsthejobdone
Yes. Nothing is gonna slow this train down.
Just accept that AI is humanity as smart phones have been the past 10 years. The next dominant species as we have been for the past 300k-3 million years depending on if u consider older hominids as human or not
Just accept that AI is humanity as smart phones have been the past 10 years. The next dominant species as we have been for the past 300k-3 million years depending on if u consider older hominids as human or not
Page 2 of 2
Popular
Back to top
Follow TigerDroppings for LSU Football News