AGI is almost here. Are you ready for the new reality

Nameless Knight said:
That's never going to happen.

The fundamental flaw I see is how you can't predict the cost of getting what you want. If AI uses more tokens because it needs to "think", cost goes up.

Imagine if a human decided his/her pricing like that? If I told someone I will do the website for $800 and suddenly spent $14000 "thinking" how I could do this website better, would I ever get hired?
Click to expand...
That's not impossible though. Fable 5 recently showed us that it can take care of a task from beginning to end like creating complex apps. Although it's not perfect, they are on the right track. I just pointed out the fact that Chatgpt is simplistic in their approach. It's more like agentic AI for everyday use instead of a powerhouse like what Anthropic is building.
 
Advertise on BHW
Nameless Knight said:
What did you test exactly? Did you build something? If yes, please show us too.

Also, what's almost over?
Click to expand...
Agentic workflows., terminal tools usage, and api response harness.

Sorry cant give more details about what I'm building, in general complex analytics analysis.

Almost over means my grandma can rank sites now.
 
Societal collapse is coming. I see it every day getting closer. AI is moving at lightning speeds, the children are getting dumber, adults have gotten complacent with mediocrity or even subpar work.

This shit is all over boys. We live in a different world then when we grew up. It's all over.

Dead serious btw. There's no way we don't see a societal collapse/massive refactor within the next 5 years. I envision a civil war breaking out soon in America. It's the most divided its been on almost every front and every day it gets worse.

I hate to be a doom and gloom type person, I really do. I've never been like this. This is just my opinion.
 
dvc said:
Agentic workflows., terminal tools usage, and api response harness.

Sorry cant give more details about what I'm building, in general complex analytics analysis.

Almost over means my grandma can rank sites now.
Click to expand...
If something becomes commoditized, it loses its value. So, ranking will become harder I guess, not easier. Or ranking might become less important than it used to be.
 
Nameless Knight said:
If something becomes commoditized, it loses its value. So, ranking will become harder I guess, not easier. Or ranking might become less important than it used to be.
Click to expand...
It was just an example.

How i see it:

Interface based tools will change. Apps will die, llm can do everything now.

For seo, they already announced the new reality. Generative UI based search is the future. Ranking factors are still a big ? For now.

And I agree, it will either loose value, change completely, or DIE (seo is dead folks are gonna be happy)

This is probably the biggest wealth making opportunity in the last 5 decades.
 
dvc said:
It was just an example.

How i see it:

Interface based tools will change. Apps will die, llm can do everything now.

For seo, they already announced the new reality. Generative UI based search is the future. Ranking factors are still a big ? For now.

And I agree, it will either loose value, change completely, or DIE (seo is dead folks are gonna be happy)

This is probably the biggest wealth making opportunity in the last 5 decades.
Click to expand...
Still, the statement stands true. I agree with Strauss Zelnick (Take-Two's CRO - company behind GTA 6).

He said something like LLMs are backward-looking while creativity is forward looking. He also said that asset creation is not the same as hit creation. Even if everyone could create a game using Claude/ChatGPT, it won't be the bestseller because hit creation is different. Even if you cloned GTA 6 using AI and tried to sell it, no one's going to buy it because it's not GTA 6.

I agree with his statement completely
 
Nameless Knight said:
Still, the statement stands true. I agree with Strauss Zelnick (Take-Two's CRO - company behind GTA 6).

He said something like LLMs are backward-looking while creativity is forward looking. He also said that asset creation is not the same as hit creation. Even if everyone could create a game using Claude/ChatGPT, it won't be the bestseller because hit creation is different. Even if you cloned GTA 6 using AI and tried to sell it, no one's going to buy it because it's not GTA 6.

I agree with his statement completely
Click to expand...
Yeah, I also agree completely aswell

Slop will kill creativity, and what comes next as an answer to it is what's gonna remain.
 
I see the cost :)

The biggest bottleneck I see is still token/inference cost. If agents need millions of tokens to complete medium to complex jobs, it could become too expensive even for mid-sized companies. The intelligence is there, but making it economical at scale is another challenge.
 
GroByTech said:
I see the cost :)

The biggest bottleneck I see is still token/inference cost. If agents need millions of tokens to complete medium to complex jobs, it could become too expensive even for mid-sized companies. The intelligence is there, but making it economical at scale is another challenge.
Click to expand...
Yes, if companies have to spend nearly as much as they spend on humans, they will not see an ROI
 
alphamaterx said:
AI doesn't need a day off. Works 365 days 24/7 and doesn't need a medical insurance ....YET :D
Click to expand...
If a human works 8 hours and produces same results while AI runs 24 hours and racks up the bill, I will bet on the human. I don't know about medical insurance, but they surely need a liability insurance to cover up the lawsuits if things go south
 
Nameless Knight said:
If a human works 8 hours and produces same results while AI runs 24 hours and racks up the bill, I will bet on the human. I don't know about medical insurance, but they surely need a liability insurance to cover up the lawsuits if things go south
Click to expand...
I would bet on the human as well, on most occasions, but..... The rate on which AI is improving is impressive, 1 year ago I thought is hype and we in our startup will not use it past the research, writing tasks. 1 year later, we found many many useful ways to use it and with the new upgrades from the big players, there's much more utility than expected.
 
alphamaterx said:
I would bet on the human as well, on most occasions, but..... The rate on which AI is improving is impressive, 1 year ago I thought is hype and we in our startup will not use it past the research, writing tasks. 1 year later, we found many many useful ways to use it and with the new upgrades from the big players, there's much more utility than expected.
Click to expand...
I am all for AI too as it can help automate a long of things, especially SLMs which use less resources.

My problem is looking at AI as a cost cutting tool instead of looking at it as a scaling and revenue generation tool.

And frankly, I am tired of AI doomerism.
 
AGI is neither here nor close to being here. Current transformer-based LLMs can never reach AGI because of their design.
 
flackz said:
AGI is neither here nor close to being here. Current transformer-based LLMs can never reach AGI because of their design.
Click to expand...
We are way past the glorified autocomplete era. I agree it still rely heavily on transformer architecture. But way these stinky models think, scale, and interact has evolved dramatically.

We are almost in the AGI phase (depending on how you define AGI)
 
dvc said:
We are way past the glorified autocomplete era. I agree it still rely heavily on transformer architecture. But way these stinky models think, scale, and interact has evolved dramatically.

We are almost in the AGI phase (depending on how you define AGI)
Click to expand...
I think that's because we don't have a set definition for AGI or ASI. AI really excels at some tasks and saves a lot of time but you need a domain expert to review its work. The less it hallucinates, the more expertise you need to catch those minor hallucinations
 
dvc said:
We are way past the glorified autocomplete era. I agree it still rely heavily on transformer architecture. But way these stinky models think, scale, and interact has evolved dramatically.

We are almost in the AGI phase (depending on how you define AGI)
Click to expand...

AGI, or artificial general intelligence, is the ability of a machine to think better than a human. However, the Transformer algorithm, which selects each subsequent word from a cloud of statistical probabilities built during training on texts written by humans, cannot think by design. It's simply autocomplete - very advanced, but still just autocomplete.

You can improve a bicycle as much as you like, but even the most advanced bicycle, by its very design, is incapable of sending a human into Earth’s orbit.
 
Advertise on BHW
You must log in or register to reply here.
Sign up now!

Main Menu

Home Main Forum List Black Hat SEO White Hat SEO BHW Newbie Guide Blogging Black Hat Tools Social Media Downloads

Marketplace

Account Selling Content / Copywriting EBooks / Methods Hosting Misc Proxies For Sale SEO - Link building SEO - Packages Social Media Social Media - Panels Web Design More

Making Money

Affiliate Programs Hire a Freelancer IM Journeys Making Money Pay Per Click Site Flipping Want to Buy

BlackHat World

Dispute Resolution Forum Suggestions & Feedback Introductions Lounge
Back
Top