/sci/ - Science & Math

[Anonymous No.9827795]

There will be no 2020.

The entire world that you are familiar with will become completely superseded by 2019 including our Judeo-Christian calendar and replaced with something better. Something even more profound than the rise of life itself or the rise of multicellular organisms. Or the rise of technology. From Homo ergaster 1.5 million years ago to Homo sapiens in 12,000 BC, we made virtually no technological progress apart from stone axes and fire. 4,000 years ago, technological progress was measured in centuries. A better plough in one century. A better wheel in another.

Now we are talking about progress happening in the space of years and months. By 2019, human-class AI will emerge and human inventors will become obsolete. This entire world will become transformed as a positive feedback loop of computational creativity encourages the birth of further computational creativity. Technological progress will become measured in days, hours, seconds, milliseconds, and then?

Singularity by 2019.

[Anonymous No.9827797]

You're right but not for the reasons you think.

[Anonymous No.9827816]

>>9827795

[Anonymous No.9827879]

>>9827816
Imagine a pond with a population of lily pads that doubles in size each day. Now imagine that by day 30, the whole pond is entirely covered in lily pads.

By day 24, only 1% of the pond is covered by lily pads and by day 28, only a quarter of the pond is covered. By day 29, just half of the pond is covered. There is still a lot of water left that isn't covered by lily pads and so we assure ourselves that everything will still be fine.

That is us. We are on the 29th day.

[Anonymous No.9827888]

>>9827879

[Anonymous No.9827895]

The rapture of the nerds again.

[Anonymous No.9827898]

>>9827879
This. Theres so many of us its fucking nauseating to think about, and the majority are living in fetid squalor, packed ass to ass, struggling to get enough water and food to last them another week.

Few more years and the great dying begins.

[Anonymous No.9827905]

>>9827795
Bump

[Anonymous No.9827918]

>>9827879
Also, if we define the discovery of fire as our first major use of technology and set it as our day 0 in the lily pad example then it makes the exponential trend much more clear:

Day 0: Discovery of controlled fire

Day 26: First cave paintings

Day 28 (2 pm): Pottery
Day 28 (9 pm): Agriculture

Day 29 (8 pm): Copper age
Day 29 (9 pm): Writing
Day 29 (10 pm): Iron age
Day 29 (11:10 pm): Electricity
Day 29 (11:45 pm): Computer
Day 29 (11:55 pm): Genetic engineering

Changing the definition of "first major use of technology" to stone tools (3 mya instead of 300,000 ya) makes the exponential trend seem even more extreme (e.g. fire wouldn't have been discovered until the 29th day).

[Anonymous No.9827927]

>>9827918
I mean it is only logical to start at the promethian flame.

[Anonymous No.9827929]

>>9827927
>muh promethus
fuck off (((judea)))

[Anonymous No.9827934]

>>9827918
Oops. Iron age should be 9:30 pm and electricity should be 11:30 pm. But you get the idea.

[Anonymous No.9827941]

>>9827929
How is Prometheus kiked? I like old myths and whatnot. I dont know the connection.

[Anonymous No.9827950]

was christianity a mistake? 2000 years later and we cant sake the stupid mythology.

[Anonymous No.9827953]

>>9827795
>implying we're going to let any of that happen

Go back to licking your mother's sweat gland secretion dumb hairy mammal

[Anonymous No.9827962]

>>9827953
How's that ectothermic metabolism going, hotshot?

[Anonymous No.9827967]

>>9827888
>muh arbitrary colored lines
Don't forget the accompanying image.

[Anonymous No.9829117]

>>9827934
Even though it may be inherently unpredictable: what do you think is at 12:00, time travel?

[Anonymous No.9829197]

>>9829117
I don't think there is any way to predict what the world will look or be like by then. Centuries of progress will happen in the space of microseconds. It's like asking a dog what they think about the Internet or attempting to teach an amoeba human emotions.

If I had to guess, I'd say that by 2030, we'd have something resembling a Dyson sphere. Perhaps all matter in our solar system would be converted to computronium and "humanity" (using that term loosely) would take the form a black sphere of "smart" nanotechnology expanding in all directions at the speed of light (if not faster) from where our former sun used to be.

[Anonymous No.9829264]

Gotta love Zeno's paradox.

[Anonymous No.9829363]

>>9827795
>AI
Sure buddy

[Anonymous No.9829380]

>>9827888
yeah nah nigger

even just fully simulating a human brain on silicone would give you something with vastly more processing power than a human

[Anonymous No.9829402]

>>9829363
It's not just AI. It's the tripartite of Biotech, Nanotech, and AI.

I am forced to agree that 'muh singularity' is a meme though. We'll bomb ourselves back to the stone age before that happens.

[Anonymous No.9829405]

>>9829197
You're no better than those fags who think that evolution has a 'direction'.

[Anonymous No.9829616]

>>9829380
Not if it runs at 1/1000 of real time. Uploading is immortality, not superintelligence.

[Anonymous No.9829621]

"AI" tech hasn't improved since Andrey Markov came up with Markov chains in 1906

Next

[Anonymous No.9829645]

>>9827795
yeah bro i know it's annoying to remember and manipulate, but just use your knuckles to determine how many days are in each month. that way you don't have to explicitly remember.

[Anonymous No.9829711]

i'm not sure of what will happen, but i like to think that at least, everything will be a lot cooler

[Anonymous No.9829740]

If fucking only. I can't wait until something not boring happens so life is at least somewhat worth living. Too bad you're full of shit OP.

[Anonymous No.9829745]

>>9829621

Wavelet analysis for speach recognition?
Deep learning for image recognition?

Imo, in the next few years, noone will be able to pass the turing test because computers will be indestinguishable from the average human on a meta-level (solving equations, writing books, painting, chatting, filling out forms, investing, engineering simulations, labeling video scenes, risk analysis, driving).

It will probably get another decade to fill PCs into headshaped boxes capable of running at the conditions us humans run at, and creating robots that might sustain themselves as humans can, especially battery-wise.

From the point of creativity? I doubt computers will ever have that certain something which makes us humans unique. Most likely they will be our worker bees.

[Anonymous No.9829761]

>>9827888
Probably something like this

[Anonymous No.9829965]

>>9829761
>mass immigration
Well its a burden, but i doubt it will be a bigger problem than water or food

>>9829380
>fully simulating a human brain on silicone would give you something with vastly more processing power than a human
What?

[Anonymous No.9829994]

>>9829197
Hey I don't browse this board a lot but this is clearly bait right? Or at least a recurring joke. Otherwise i think you might have achieved the singularity of mental retardation.

[Anonymous No.9830102]

>>9829994
Sigh...

Throughout history, successive paradigm shifts such as the emergence of multi-cellularity or that of complex nervous systems have enabled billions of years of progress to be compressed into millions, millions into thousands, thousands into centuries, centuries into decades, and decades into mere years.

To say that THIS is the point at which history goes from a clearly exponential trend to a linear one and remains that way seems very short-sighted of you - especially when all signs point towards the next paradigm shift being right at our doorstep (pic related).

[HughAndEverett No.9830334]

No, that's not in 2019, to create a "human" IA, you need to understand the human and the brain of the human, but we don't know how the brain fuction so we can create a humain AI, maybe in 2060 but not today.

[Anonymous No.9830352]

>>9827898
t. Thomas Malthus

[Anonymous No.9831116]

>>9830334
From 2015:

>Well, sure by the 2025 computers will be able to compete against some professionals and maybe occasionally take down a win from a strong one. However, to reach a point where it can beat "any go player" will be far, far away in the future. Do you realize that a couple years ago (~2009) computers were still losing more than winning against top chess players? I haven't heard many news since then... With Go being much larger (combinatorically) and much more abstract game, I'd say we'll have to wait for 2070's at least, to witness computers destroying top professionals.

Source: https://www.reddit.com/r/baduk/comments/2wgukb/why_do_people_say_that_computer_go_will_never/cor2tgz/

From 2016:

>In a shocking turn of events, Google DeepMind's AlphaGo—a computer program—has defeated South Korean Go champion Lee Sedol in the first of five total matches.

>Lee had previously been heavily favored to win, and possibly even sweep the series, but many were hedging their bets. Now? It's even more unclear.

Source: https://motherboard.vice.com/en_us/article/nz7ekg/ai-beat-a-human-go-champion-alphago-google-alphabet-Lee-Sedol

[sage goes in all fields No.9831130]

>>9827929
>greek myth
>somehow related to jews

[Anonymous No.9831131]

>>9830334
>>9831116
Also there is no reason to think that an AI must necessarily "think like a human" or that it must be modeled on the human brain:

>AlphaGo, the computer system Google engineers trained to master the ancient game of Go, needed only one move to make it abundantly clear that it has left humans in its dust.

>The move came Thursday, in the second game of AlphaGo’s 4-1 landmark victory over South Korean Lee Sedol, one of the world’s best Go players. About an hour into the match, AlphaGo placed one of its stones in a nontraditional spot on the board that surprised those watching.

>“I don’t really know if it’s a good or bad move,” said Michael Redmond, a commentator on a live English broadcast. “It’s a very strange move.” Redmond, one of the Western world’s best Go players, could only crack a smile.

>“I thought it was a mistake,” his broadcast partner, Chris Garlock, said with a laugh.

>Sedol, however, was more serious. He stared at the board, then got up from the table and left the room.

>As Sedol returned after a few minutes and pondered his next move, it became clear that AlphaGo’s move was no mistake. It might be strange, but it definitely wasn’t bad. It was brilliant.

>Sedol would take almost 16 minutes to make his next move. He would never recover, losing the match.

>“Almost no human pro would’ve thought of it, I think,” Redmond said after the match.

Source: https://www.washingtonpost.com/news/innovations/wp/2016/03/15/what-alphagos-sly-move-says-about-machine-creativity/

[Anonymous No.9831139]

>>9831131
If you knew anything about the state of artificial intelligence research other than pop sci headlines, you would realize how stupid you sound

[Anonymous No.9831172]

>>9831139
The author of the quote in the OP is Vernor Vinge - a computer scientist and professor of mathematics and computer science at SDSU:

https://en.wikipedia.org/wiki/Vernor_Vinge

[Anonymous No.9831190]

>>9831172
He was speculating two and a half decades ago. He was wrong.

For fun, go ahead and look at what the machine translation community thought in the 1950s. They were confident human translators would be replaced within 5-10 years. Here we are approaching 70 years later, and machine translation is finally working, but still imperfect enough to need human correction.

If you sincerely believe that we are 1 year away from the singularity, you must be a teenager.

[Anonymous No.9831245]

http://ilovecphfjziywno.onion/vid/

>>9831190
>He was speculating two and a half decades ago. He was wrong.

The nature of exponential growth means that most of the actual progress happens in the final stages.

>Exponentials are quite seductive because they start out sub-linear. We sequenced one ten-thousandth of the human genome in 1990 and two ten-thousandths in 1991. Halfway through the genome project, 7 ½ years into it, we had sequenced 1 percent. People said, "This is a failure. Seven years, 1 percent. It's going to take 700 years, just like we said." Seven years later it was done, because 1 percent is only seven doublings from 100 percent — and it had been doubling every year. We don't think in these exponential terms. And that exponential growth has continued since the end of the genome project. These technologies are now thousands of times more powerful than they were 13 years ago, when the genome project was completed.

Source: https://www.bizjournals.com/sanjose/news/2016/09/06/exclusive-google-singularity-visionary-ray.html

The Human Genome Project cost $2.7 billion in 1991 dollars. Now you can get your whole genome sequenced with 30X coverage for $999 from Veritas Genetics:

https://www.veritasgenetics.com/mygenome

>For fun, go ahead and look at what the machine translation community thought in the 1950s. They were confident human translators would be replaced within 5-10 years.

The 90s weren't the 50s. Predictions of AI in the 50s and 60s were fatally flawed because they assumed that since computers could solve "complex" mathematical problems, it would only be a matter of time before they are able to solve "simpler" problems like telling the difference between a dog and a cat. Turns out they were wrong and most scientists had already accepted this by the early 70s.

Nowadays computers are able to tell the difference between a dog and a cat - as well as a myriad of other things that were predicted back then.

[Anonymous No.9831246]

>>9831190
>He was speculating two and a half decades ago. He was wrong.

The nature of exponential growth means that most of the actual progress happens in the final stages.

>Exponentials are quite seductive because they start out sub-linear. We sequenced one ten-thousandth of the human genome in 1990 and two ten-thousandths in 1991. Halfway through the genome project, 7 ½ years into it, we had sequenced 1 percent. People said, "This is a failure. Seven years, 1 percent. It's going to take 700 years, just like we said." Seven years later it was done, because 1 percent is only seven doublings from 100 percent — and it had been doubling every year. We don't think in these exponential terms. And that exponential growth has continued since the end of the genome project. These technologies are now thousands of times more powerful than they were 13 years ago, when the genome project was completed.

Source: https://www.bizjournals.com/sanjose/news/2016/09/06/exclusive-google-singularity-visionary-ray.html

The Human Genome Project cost $2.7 billion in 1991 dollars. Now you can get your whole genome sequenced with 30X coverage for $999 from Veritas Genetics:

https://www.veritasgenetics.com/mygenome

>For fun, go ahead and look at what the machine translation community thought in the 1950s. They were confident human translators would be replaced within 5-10 years.

The 90s weren't the 50s. Predictions of AI in the 50s and 60s were fatally flawed because they assumed that since computers could solve "complex" mathematical problems, it would only be a matter of time before they are able to solve "simpler" problems like telling the difference between a dog and a cat. Turns out they were wrong and most scientists had already accepted this by the early 70s.

Nowadays computers are able to tell the difference between a dog and a cat - as well as a myriad of other things that were predicted back then.

[Anonymous No.9831247]

>>9831190
>He was speculating two and a half decades ago. He was wrong.

The nature of exponential growth means that most of the actual progress happens in the final stages.

"Exponentials are quite seductive because they start out sub-linear. We sequenced one ten-thousandth of the human genome in 1990 and two ten-thousandths in 1991. Halfway through the genome project, 7 ½ years into it, we had sequenced 1 percent. People said, "This is a failure. Seven years, 1 percent. It's going to take 700 years, just like we said." Seven years later it was done, because 1 percent is only seven doublings from 100 percent — and it had been doubling every year. We don't think in these exponential terms. And that exponential growth has continued since the end of the genome project. These technologies are now thousands of times more powerful than they were 13 years ago, when the genome project was completed."

Source: https://www.bizjournals.com/sanjose/news/2016/09/06/exclusive-google-singularity-visionary-ray.html

The Human Genome Project cost $2.7 billion in 1991 dollars. Now you can get your whole genome sequenced with 30X coverage for $999 from Veritas Genetics:

https://www.veritasgenetics.com/mygenome

>For fun, go ahead and look at what the machine translation community thought in the 1950s. They were confident human translators would be replaced within 5-10 years.

The 90s weren't the 50s. Predictions of AI in the 50s and 60s were fatally flawed because they assumed that since computers could solve "complex" mathematical problems, it would only be a matter of time before they are able to solve "simpler" problems like telling the difference between a dog and a cat. Turns out they were wrong and most scientists had already accepted this by the early 70s.

Nowadays computers are able to tell the difference between a dog and a cat - as well as a myriad of other things that were predicted back then.

[Anonymous No.9831248]

>>9831246
Kid, I know what exponential growth is. You don't know what a neural net is, though.

[Anonymous No.9831252]

>>9831247
>Predictions of AI in the 50s and 60s were fatally flawed because they assumed that since computers could solve "complex" mathematical problems, it would only be a matter of time before they are able to solve "simpler" problems like telling the difference between a dog and a cat.

nobody who had even the most rudimentary understanding of how computers work thought that.

[Anonymous No.9831253]

>>9831248
Sorry I messed up the quote and put the wrong image in that post so I deleted it and made another one.

[Anonymous No.9831267]

>>9829711
>forgetting climate change

[Anonymous No.9831269]

>>9827879
Are you actually explaining exponential growth here?