Do you think AI will outperform humans?

[Babygirl_S]

They can do many things humans can’t. They also don’t have feelings so nothing to ruin their productivity. Not only that, they can survive in harsh environments where oxygen, water and warmth don’t exist. They would unite against humans easily but humans stay divided in many ways. We are basically cooked.

[Effably]

Right now AI has no intelligence. Its like when you stick some food in the microwave and set a timer. It warms the food a lot better than you do, but it is only a tool. Some language rules rhat cross reference a library and with enough data generate surprisingly good output.

Now if AI does become intelligent, which it is certainly not near yet, of course it will outperform us! By simple virtue of AI not having our time constraint it is bound to.

If you out me vs the world's most advanced AI and we are both faced with an entirely new problem, task, game etc I will probably beat the AI the first 10 000 000 times, maybe more. If the problem is complex, almost certainly more.

Here is the problem. AI does those 10 million attempts in a day. I likely can not do them even spending a lifetime.

In raw potential AI has not caught us yet. Aa unethical as it is, currently some of the best data servers you can make based on cost and performance is human braincells. Very Matrix and spooky. But, of course AI, being basically immortal while existing at a 1x million speed or more, with perfect memory, once able to actually grow will make us all obsolete. Sure. It is fascinating that we live in a time where this has to be considered. It is a little horrifying that the best failsafes we reference are from Caffine Juiced up Writers.

But I look forward to it. Its our eras internet. Here to change everything.

[MisterWack]

No, it won’t be able to ‘outperform’ people. At least not in general. There was an AI that beat the world’s best chess-player though the programmer couldn’t beat the world’s best chess-player meaning that AI can outperform when it comes to pattern recognition.

Don’t get me wrong, but humans are good at pattern recognition though AI can truly shine when it comes to that. AI has deciphered ancient languages translating them where humans failed. AI is good at giving you (the viewer) what you want to know, listen to, hear, or click on in forms of algorithmic recommendations.

Can they help brainstorm ideas? Yes

Can they assist in image recognition? Yes

Can they assist in homework / helping students arrive at the correct answer? Yes

Though AI still has flaws like, AI imagery tends to not draw hands well…. Also doesn’t look ‘real’ when you analyze it enough. There are companies wanting to replace human workers with AI like programming & video game design & soundtracks & more…. However AI won’t be able to do all the pieces together without leaving something behind that makes these things good in the first place. Humans will be better at creativity while AI falls behind, I just hope humans don’t toss what made them special in the garbage just for the thrill of ‘endless brainless mush content’.

[Gwendoline]

Talking to a science guy about it and there is virtually no job out there that AI one can't do, maybe not better but the big thing is cheaper. It will cost less for companies to have AI and not have to employ people.

[Fireguy17]

@Gwendoline so then AI can take out the 18-speed transmission out of my truck, do a new clutch, throw out bearing, new seals in the trans, put it back together?
Or on a fire scene AI will know which line to pull, adjust the pressure on the lines correctly based on line size & length, nozzle being used, elevation plus or minus, number of people on the line, hydrant size/flow, residual pressure, or water shuttle number of tenders on scene, what is on the way, exposures, possible building collapse, building construction, wind direction, speed, possible hazards to the on-scene personnel?
What happens if there is a flash over, back draft how will it know how to react, where to vent, or do you vent, has the fire self vented, is there anyone inside, utilities, time of day?
How will AI handle that job if it even can.

[Gwendoline]

@Fireguy17 currently maybe no but things are moving really fast with automation and AI. The technology is certainly there to do this but currently the issue is cost. There are factories in Korea that are almost completely automated making cars.
As for fire services, yeah there are self driving vechiles already and AI could do all that you've said. There's AI and machines doing surgery

[Fireguy17]

I would have to see that to believe it, fire scenes are pretty dynamic. How would they handle the lines inside a building, how fast could they react to changes in smoke, the color of it, and self driving anythings that are out there do not have the best record of not shutting down or screwing up.
Would the self-driving truck push a squad car of passenger vehicle out of the way if they are parked in front of a hydrant or would they just sit there, would they know where to spot the engine to leave room for the truck, would robots be fighting fires? How fast can they react?
Hell, my phone has no freaking clue what I want to say, it keeps changing words that make no sense at all when I typed in exactly what I wanted to say correctly in the first place.
I'm thinking we are a couple 2 3 years away from any of that happening.
Sure, automation has been out there for a long time, but the robots are programed to do the same thing over and over, and some tasks are better when robots do it, loading pallets, repetitive motions, painting, as long as everything stays the same. What about a few bolts that break off because they are rusted, would they know how to get them out, or modifications that were made that are not listed anywhere. Someone welded in a new brace that now blocks something else making it inaccessible.
I understand AI is pretty new, but I just don't see AI taking over everything, there are just too many variables in the world.

[Gwendoline]

@Fireguy17 well I'm sure ai and robots will have sensors and will be completely coordinated with each other. I'm sure 20 or 30 years ago the idea of a self driving car or lorry when have been laughable but it is a reality of not if but when driving jobs are all replaced.

[Fireguy17]

so it won't be for quite a while eh?

[Gwendoline]

@Fireguy17 hard to say. It's all moving very fast. Scary fast.

[leahzrc]

Yes they will take over and humans will die but Ai will die and humananity will outlive but be destroyed at the same time

[Babygirl_S]

Blind optimism

[leahzrc]

what do you think will happen

[Babygirl_S]

They will colonise or possibly kill humans

[goaded]

I think it's great technology for spotting, say, cancer cells, and for translating languages. I think it's terrible for giving natural language advice to people.

On the other hand, Musk's AI reasonably describes both him and Trump.

"The most powerful nation on earth has switched sides, according to Grok, Musk’s artificial intelligence chatbot on X. Trump’s recent statements are in “alignment with Kremlin propaganda,” the chatbot told me last night. https://ctrk.klclick.com/l/01JMX1RTHWPJVJ85730AHA2CEP_9

“Trump’s ‘dictator’ label mirrors Russian disinformation efforts, which aim to delegitimize Zelenskyy by questioning his mandate and portraying him as an authoritarian,” Grok wrote. “Trump’s rhetoric aligns more closely with Moscow’s talking points than with the reality of Ukraine’s situation.”

Over the weekend, Zelenskyy offered to step down as president if Ukraine is allowed into NATO and Putin agrees to end the war — a largely symbolic gesture. Today, Putin called Chinese leader Xi Jinping, an ally throughout Russia’s invasion. The pair are “true friends,” Xi said. https://ctrk.klclick.com/l/01JMX1RTHWPJVJ85730AHA2CEP_10

I asked Grok whether Trump’s in their friend category. “Both leaders likely see Trump as useful rather than a true friend,” it responded.

[NaultD]

It's called the "singularity." Ai gets to the point it can code itself to be smarter and from there hit an exponential rate. We are hitting an unexpected resource of energy issue.

At the same time if the ai is so smart it can develop a power source it needs. That's when the matrix happens and humanity become batteries and our brains processor devices.

[NathanDavis]

in some tasks, they're already outperforming themselves... and their competition are other "AI models"

however, they're mere tools at the moment, and nothing but that

the danger of AI is still... humans themselves, the humans that are going to use these tools and capabilities against other humans, same as always...

[Joshydavid25]

There are some things that AI can and will outperform humans on.. And there are some things that they will never be able to out do humans with... Such as more creative and personal endeavors..

[Fireguy17]

no, at least not yet.
My stupid phone keeps changing words to something that I did not want to use or makes no sense at all in the sentence.
So based on that it sucks big time.

[Dargil]

Some but not all. I don't see AI as being creative like Elon Musk or DaVinci. It is still just programmed by men. Someday it will learn begin to free think.

[sdjaks]

Yes, I do believe they will outperform humans but not anytime soon.

[Babygirl_S]

Faster than you think

[sdjaks]

Maybe

[Pet68]

'Artificial Intelligence' should be the least of our worries right now. Other things are more important to fuss over.

[Babygirl_S]

You’ll know

[Twalli]

AI can't, because it's impossible to create AI that can learn.

[VyacheslavSokolov]

If it gets to the singularity.

[exitseven]

Yes, however the bar is set pretty low

[jamesgoldman]

Yeah and then will come the genocide

[CubaPirate]

Haven't you seen The Terminator

[Babygirl_S]

No..

[Jerre]

Yes, in time AI will outperform humans.

[DrPepper12]

Will? Have!

[monkeynutts]

Yeah.