How accurate are Ray Kurzweil's predictions?

• Virtual artists—creative computers capable of making their own art and music—emerge in all fields of the arts.

NOPE -- The Culture we are yet not

• Massively parallel neural nets and genetic algorithms are in wide use.

NOT QUITE

Claim one. I mostly object. I study the AI field a lot, and Ian Goodfellow invented GANs in 2014, and if you watch Two Minute Paper videos on Youtube - you can see how they learn to generate fake artwork, faces, increase photo resolution, add styles, videos from line drawings, text-to-image!!!!, image colorization from B&W reallllly extra good!!, make faces smile! magic!, etc!, it's the bleeding edge and amazing, mind-dropping. A painting made by AI was recently just sold for a huge amount of money. Music remains much more dry but they are generating piano music, lip sync real time generate audio to a face with no sound, even stories using Sequence Prediction that model a given dataset to look like. As for in all fields of the arts, I would say so, because as you can read on articles and AI Papers - they are plowing out so many types of AI neural network variants and they are being applied to so many crazy things! Like detecting cancer, or making someone else dance to a pose, etc.

!!!! https://gizmodo.com/...-int-1828413827

See mine I made attached, it is so cool.

 6556865868.PNG   819.11KB   0 downloads

 generated one.PNG   367.83KB   1 downloads

 2.PNG   396.91KB   0 downloads

>>> !!!!!! GROWING PLUS MORPHING CELEB FACES

As for the second claim, not arguing it much at all, but, not sure if Google is using them much but before we know it our cellphones may be using their cloud's genetic algorithm, or a GAN or something better. The rate of the AI field is astounding. Definitely not far off from happening or being the case at present. And, the amount of AI scientists using them is a lot, so in a sense it is widespread.

Edited by Dream Big, 19 January 2019 - 07:30 AM.

To further that, text to video:

Text to big resolution bird image:

(refresh the page and try again if it doesn't pop up your image in like 7 seconds (maybe))

https://drawingbot.azurewebsites.net/

Read about it here: https://www.ailab.mi...41-0118f414bd78

Edited by Dream Big, 20 January 2019 - 04:19 PM.

For beginners to AI / AGI, you will get a beyond-well drive if you check out all the content in the 3 links please:

https://openai.com/

https://openai.com/b...se-transformer/

https://www.youtube..../keeroyz/videos

Specifically, GPT-2, Musenet, etc on Open AI, are prime areas.

Edited by Dream Big, 21 May 2019 - 02:47 AM.

 

• The computational capacity of a $4,000 computing device (in 1999 dollars) is approximately equal to the computational capability of the human brain (20 quadrillion calculations per second).
• The summed computational powers of all computers is comparable to the total brainpower of the human race.
• Computers are embedded everywhere in the environment (inside of furniture, jewelry, walls, clothing, etc.).
• People experience 3-D virtual reality through glasses and contact lenses that beam images directly to their retinas (retinal display). Coupled with an auditory source (headphones), users can remotely communicate with other people and access the Internet.
• These special glasses and contact lenses can deliver "augmented reality" and "virtual reality" in three different ways. First, they can project "heads-up-displays" (HUDs) across the user's field of vision, superimposing images that stay in place in the environment regardless of the user's perspective or orientation. Second, virtual objects or people could be rendered in fixed locations by the glasses, so when the user's eyes look elsewhere, the objects appear to stay in their places. Third, the devices could block out the "real" world entirely and fully immerse the user in a virtual reality environment.
• People communicate with their computers via two-way speech and gestures instead of with keyboards. Furthermore, most of this interaction occurs through computerized assistants with different personalities that the user can select or customize. Dealing with computers thus becomes more and more like dealing with a human being.
• Most business transactions or information inquiries involve dealing with a simulated person.
• Most people own more than one PC, though the concept of what a "computer" is has changed considerably: Computers are no longer limited in design to laptops or CPUs contained in a large box connected to a monitor. Instead, devices with computer capabilities come in all sorts of unexpected shapes and sizes.
• Cables connecting computers and peripherals have almost completely disappeared.
• Rotating computer hard drives are no longer used.
• Three-dimensional nanotube lattices are the dominant computing substrate.
• Massively parallel neural nets and genetic algorithms are in wide use.
• Destructive scans of the brain and noninvasive brain scans have allowed scientists to understand the brain much better. The algorithms that allow the relatively small genetic code of the brain to construct a much more complex organ are being transferred into computer neural nets.
• Pinhead-sized cameras are everywhere.
• Nanotechnology is more capable and is in use for specialized applications, yet it has not yet made it into the mainstream. "Nanoengineered machines" begin to be used in manufacturing.
• Thin, lightweight, handheld displays with very high resolutions are the preferred means for viewing documents. The aforementioned computer eyeglasses and contact lenses are also used for this same purpose, and all download the information wirelessly.
• Computers have made paper books and documents almost completely obsolete.
• Most learning is accomplished through intelligent, adaptive courseware presented by computer-simulated teachers. In the learning process, human adults fill the counselor and mentor roles instead of being academic instructors. These assistants are often not physically present, and help students remotely.
• Students still learn together and socialize, though this is often done remotely via computers.
• All students have access to computers.
• Most human workers spend the majority of their time acquiring new skills and knowledge.
• Blind people wear special glasses that interpret the real world for them through speech. Sighted people also use these glasses to amplify their own abilities.
• Retinal and neural implants also exist, but are in limited use because they are less useful.
• Deaf people use special glasses that convert speech into text or signs, and music into images or tactile sensations. Cochlear and other implants are also widely used.
• People with spinal cord injuries can walk and climb steps using computer-controlled nerve stimulation and exoskeletal robotic walkers.
• Computers are also found inside of some humans in the form of cybernetic implants. These are most commonly used by disabled people to regain normal physical faculties (i.e. - Retinal implants allow the blind to see and spinal implants coupled with mechanical legs allow the paralyzed to walk).
• Language translating machines are of much higher quality, and are routinely used in conversations.
• Effective language technologies (natural language processing, speech recognition, speech synthesis) exist
• Access to the Internet is completely wireless and provided by wearable or implanted computers.
• People are able to wirelessly access the Internet at all times from almost anywhere
• Devices that deliver sensations to the skin surface of their users (i.e.--tight body suits and gloves) are also sometimes used in virtual reality to complete the experience. "Virtual sex"—in which two people are able to have sex with each other through virtual reality, or in which a human can have sex with a "simulated" partner that only exists on a computer—becomes a reality.
• Just as visual- and auditory virtual reality have come of age, haptic technology has fully matured and is completely convincing, yet requires the user to enter a V.R. booth. It is commonly used for computer sex and remote medical examinations. It is the preferred sexual medium since it is safe and enhances the experience.
• Worldwide economic growth has continued. There has not been a global economic collapse.
• The vast majority of business interactions occur between humans and simulated retailers, or between a human's virtual personal assistant and a simulated retailer.
• Household robots are ubiquitous and reliable.
• Computers do most of the vehicle driving—-humans are in fact prohibited from driving on highways unassisted. Furthermore, when humans do take over the wheel, the onboard computer system constantly monitors their actions and takes control whenever the human drives recklessly. As a result, there are very few transportation accidents.
• Most roads now have automated driving systems—networks of monitoring and communication devices that allow computer-controlled automobiles to safely navigate.
• Prototype personal flying vehicles using microflaps exist. They are also primarily computer-controlled.
• Humans are beginning to have deep relationships with automated personalities, which hold some advantages over human partners. The depth of some computer personalities convinces some people that they should be accorded more rights.
• While a growing number of humans believe that their computers and the simulated personalities they interact with are intelligent to the point of human-level consciousness, experts dismiss the possibility that any could pass the Turing Test.
• Human-robot relationships begin as simulated personalities become more convincing.
• Interaction with virtual personalities becomes a primary interface
• Public places and workplaces are ubiquitously monitored to prevent violence and all actions are recorded permanently. Personal privacy is a major political issue, and some people protect themselves with unbreakable computer codes.
• The basic needs of the underclass are met. (Not specified if this pertains only to the developed world or to all countries)
• Virtual artists—creative computers capable of making their own art and music—emerge in all fields of the arts.

 

 

We have arrived at the end of 2019 and the list seems to be as accurate as it has been in the last couple of years.

Self-driving cars are here but not widespread. I think the adoption would be quicker if there were designated routes/lanes for self-driving cars.

Still having trouble finding the cost of 20 petaflop computers (processors). I pretty sure it is getting closer to $4,000.

I think the biggest miss is the nanotube computer chips.

Another miss is the virtual learning. Vested interests (public schools) are a drag on wider adoption, IMO, but there are A LOT of options out there for self-learners.

The haptic virtual reality adoption has not materialized, but other forms are widespread.

I guess the biggest theme I see is that virtual reality is being adopted, but not so much for the "enlightened" purposes most forecasters envisioned. It has mainly been used for gaming, entertainment, etc... and not so much for discovery, science, etc...

Edited by Mind, 30 December 2019 - 08:22 PM.

Ray seems to have been quite accurate with his prediction on virtual and augmented reality but got the form factor and resolution wrong. As you said, it's not as widespread yet, but sales of Playstation VR, PCVR, and percentage of steam users has been roughly doubling each year since their release in 2016. If current trends continue, it'll probably be quite mainstream within 3-4 years. 

The capabilities of Augmented Reality will also be greatly improved with the release of Qualcomm's Snapdragon XR2 5G

Most business transactions or information inquiries will involve dealing with a simulated person.

 

https://youtu.be/D5VN56jQMWM

We might be close to this...

Edited by Matt, 30 December 2019 - 11:01 AM.

The 2019 predictions, some are trivial while others are actual game changers. It's clear the trend is all about more data/compute/manipulation. Cooperation of agents. Internet nervous system. Wireless communication. More data. Giving jobs to related humans to offload work. And this is all happening. Electricity>computer>data>AI>nanobots. The data tech builds on itself, that's why small fast phones that do lots are sold so much and used. The Hutter Prize shows AI is a Lossless Compression problem (like GANs). AGI transforms data self-recursively, extra context is the key here. It extracts free energy/knowledge by compressing. It can generate realistic/quality data better. You seen above I linked magical links, yes, the AI is really strong, biology and etc fields depend on context/wisdom, you will need to use a computer/lots of Big Diverse Data to do better stem cell research.

Since I'm here, I'll mention we are like water in a river, flowing. We aren't lucky to have the physics we have, we have no other choice. We are predictable. AI works exactly by this, big data leads to exponential wisdom because there is only so many different patterns. All words in a dictionary describe each other. So do humans. You are a part of Earth, a global not local sum. You can reach anyone just a few steps away indirectly because its a small world network.

Edited by TheGene, 09 January 2020 - 11:50 PM.

I say bravo to the predictions they are true. The computer ones are more important and more is happening there.

Also these are important, building your brain and your team OF brains:

• All students have access to computers.
• Most human workers spend the majority of their time acquiring new skills and knowledge.
• Students still learn together and socialize, though this is often done remotely via computers.

I noticed Ray mentions no more cables showing, no more rotating hard drives, no real books, etc, and we are nearing that, but its more efficient still and there's always the bugs hanging around the farm - there's only a majority that will adopt this way. So we're close/ are there on that 1. The only thing off and the only thing that ACTUALLY matters on the list (besides AI, yeah all others are weak breakthroughs!!) is nanotech, our bio sucks, but that's where big data AI comes in - EASypeasy work for them nets. It's a workforce expansion thing of more computer, more data, more arms, which lead to ever more. They will soak up data off your skin literally. We do have phone/laptop pin sized cameras everywhere and everyone uses massively parallel neural network on Google/phones or at least the computer science field does who cares about laypeople anyhow!! The VR never took off but it was impractical/not important anyway.

This one I like I wonder how better our images are of the brain. I remember Ray said some stats on it, but I never researched what 2019 ended with, only like 2013. Maybe we can see axons now or something in MRI if you get me. Anyone got stats?

• Destructive scans of the brain and noninvasive brain scans have allowed scientists to understand the brain much better. The algorithms that allow the relatively small genetic code of the brain to construct a much more complex organ are being transferred into computer neural nets.

Edited by TheGene, 11 January 2020 - 04:12 AM.

 

 

Still having trouble finding the cost of 20 petaflop computers (processors). I pretty sure it is getting closer to $4,000.

 

Makes me sad that we are very very fram from this.

Unfortunately one of Kurzweil's most important predictions, if not the most important of all, which is the advance in computational performance, is lagging behind big time. 

If you see the Top 500 Supercomputer list, updated November 2019:

https://www.top500.org/lists/2019/11/

We see that only 7 computers in the world break the 20 Petaflop performance barrier. Trinity, the 7th most powerful supercomputer and the last to make the cut into the 20 petaflop range cost $174 million back in 2015 to be built (https://en.wikipedia...(supercomputer)). Even if we assume extreme deflation since then, it would still cost thousands of times more than $4,000.

I remember saying more than 10 years ago here that the only thing that really mattered and that would be a deal breaker for the Singularity etc would be a radical slowing of the computational performance advances. And that's precisely what has happened in the 10 years since.

I really hope AI picks up in the coming decades and saves the party but i have grown to be much more pessimistic in this last decade. Hope things speed up from now on and in 2029 i'm here with my hopes back up.

You are correct about the cost of supercomputers. The top on the list right now is Summit. 148 petaflops and cost (best case scenario) $400,000,000. That is $2.7 million per petaflop. No where near $4000 per 20 petaflop.

You are correct about the cost of supercomputers. The top on the list right now is Summit. 148 petaflops and cost (best case scenario) $400,000,000. That is $2.7 million per petaflop. No where near $4000 per 20 petaflop.

Mark Bahner estimated last month:

"If I go to the Wikipedia article "FLOPS" (floating operations per second), I find data that can be graphed as shown below. Note that the y-axis is logarithmic. That means that a straight line shows the cost in dollars per gigaflop of computer power is going down exponentially.

A linear regression line through the data indicates that the cost in dollars per gigaflop has declined (and will decline) by approximately 11 orders of magnitude in the 42 years from 1982 ($100 million per gigaflop) to 2024 ($0.001 per gigaflop). If the human brain is approximately 10 petaflops (10 million gigaflops), then the cost for a computer that can perform the same number of calculations as a human brain in 2024 will be approximately $10,000 (i.e., 10 million gigaflops times $0.001 per g"

https://markbahner.t...moores-law.html

1. An intriguing thread, my only problem with technological advancement is they assume that they the greate the technological advancement the greater your quality of life, which obviously isn't the case with excess smart-phone usage. Just because we can have absolute technology augmentation, doesn't implictly mean we should. I'd really like to see CRISPR-CAS9 merge with AI, so we can start curing genetic illnesses at a ever increasing rate. As for immortality, it's unlikely, I'll be happy if I get to live to 200 years of age but this seems outlandish, sense we still need more revolutions in technological advancement and things seem to be increasingly slow as of late. Perhaps, bio-hacking underground may need to be the new norm to really speed up the experimentation process.

 

Another miss is the virtual learning. Vested interests (public schools) are a drag on wider adoption, IMO, but there are A LOT of options out there for self-learners.

It's ironic how things change due to world events. Back in Dec 2019, virtual learning and remote work were still lagging in terms of adoption. For good or ill, now we have huge society wide adoption of virtual learning and remote work due to world events. Not in the organic way that Kurzweil intended of course, but interesting nonetheless.

In terms of processing power, Intel's Chief Architect had an interesting presentation on their roadmap. A few interesting points are the huge gains in compute demands in AI at the 07:30 mark (which is going at a faster than Moore's Law pace). The slide around 12:55 or so shows a roadmap at what Intel sees may lead to a 50x increase in transistor density. On the software optimization side, the video at around 11:35 shows just how much software matters and not just hardware - showing a speed up of up to 63,000x. There are a bunch of charts and graphs that mention cost but I didn't spend enough time analyzing them to compare to Kurzweil's predictions.
 

Edited by Sdescon, 03 September 2020 - 06:50 PM.

 

Still having trouble finding the cost of 20 petaflop computers (processors). I pretty sure it is getting closer to $4,000.

 

 

 

I think he says 1999 dollars,  that's $1.56 dollars today for every $1 in 1999.   That's $6240 today's dollar.

Also he doesn't say precision of the calculations.   The rtx 3080 can do

 

on paper the RTX 3080 looks even faster than what Jensen claimed. 29.8 TFLOPS of FP32 shader performance, 238 TFLOPS of FP16 Tensor core performance, and 58 TFLOPS of ray tracing performance

https://www.tomshard...rything-we-know

 

476 TOPs int 8

https://www.hardware...res-and-rtx-io/

Previous nvidia cards could do twice as many int 4 operations as int8, so that would be 952 TOPs of int4.

The rtx 3080 will be sold soon at $699.   

For $6000 you can get nearly 10 which would be about 10 Peta ops of int4.  Or around 10 Quadrillion calculations per second.  If talking about petaflops, around 2 petaflops of half precision(16bit float).   But Kurzweil does not say integer or floating calculation, and there is research on the use of discrete synapses with limited precision.

That said only a small fraction of an nvidia card is dedicated to tensor cores.   A fully dedicated AI processor should be able to do even more calculations for the same price. 

IIRC, human synapses have been said to have less than 5 bits of precision

https://www.telegrap...ly-thought.html

Edited by Castiel, 11 September 2020 - 05:25 PM.

I was just watching an older interview with Palmer Luckey (founder of Oculus VR). 

https://youtu.be/op2yes0gd6w?t=1489

Interesting that he says in the clip that the Oculus VR HMD that he put together could have been created years earlier in 2009 with the components that were available then. If work would have been done by that time, VR would have probably been mass market by 2019 as Kurzweil predicted. 

Although, adoption of VR is now speeding up. It's been roughly on an exponential growth across platforms for a few years with only a slow down in the past few months in anticipation of new hardware coming from several companies.

Device like Oculus Quest 2 will drive fast adoption. If feels like the technology is at a tipping point and so Kurzweil was pretty accurate here.
 

I was just watching an older interview with Palmer Luckey (founder of Oculus VR). 

 

https://youtu.be/op2yes0gd6w?t=1489

 

Interesting that he says in the clip that the Oculus VR HMD that he put together could have been created years earlier in 2009 with the components that were available then. If work would have been done by that time, VR would have probably been mass market by 2019 as Kurzweil predicted. 

 

Although, adoption of VR is now speeding up. It's been roughly on an exponential growth across platforms for a few years with only a slow down in the past few months in anticipation of new hardware coming from several companies.

 

Device like Oculus Quest 2 will drive fast adoption. If feels like the technology is at a tipping point and so Kurzweil was pretty accurate here.
 

It is interesting. I personally don't know anyone who owns an Oculus, although I hear good reviews. I think it is too expensive for most people in my orbit. Perhaps there are a lot of gamers in my community who own one, but because they tend to stay at home, I just don't interact with them too often.

It's getting a lot cheaper now. Oculus Quest 2 is only $299 and it's an incredibly good headset. Just released a few weeks ago and has an XR2 Chip, almost 4K screen, 6GB ram and 64 GB / 256 GB SSD. It functions as a standalone device but also can be connected to a PC with Oculus link or wireless PCVR.

It's selling out everywhere and I expect that VR to really start taking off now. It definitely feels more consumer friendly and just good enough now for most people.

The headset is insanely good value.

Edited by Matt, 24 November 2020 - 07:33 PM.

It's getting a lot cheaper now. Oculus Quest 2 is only $299 and it's an incredibly good headset. Just released a few weeks ago and has an XR2 Chip, almost 4K screen, 6GB ram and 64 GB / 256 GB SSD. It functions as a standalone device but also can be connected to a PC with Oculus link or wireless PCVR.

 

It's selling out everywhere and I expect that VR to really start taking off now. It definitely feels more consumer friendly and just good enough now for most people.

 

The headset is insanely good value.

 

That is a good deal. I had no clue it was so cheap.

It'd be cool to combine that with google street - if its not already done. Perhaps remote travelling will become a thing, like by putting 3D cameras on remote robots that can walk anywhere and on self driving cars. Like you tell the car to go somewhere, the robots gets out of the car and goes wherever you want. Better have a good compression algorithm.

The end of another year. Kurzweil's predictions have been good, but many are obviously slow to develop.

Nanotube computer chips are not here yet, but chip companies continue to shrink the feature size (TMSC is planning for 3nm soon), so it is kind-of a good prediction.

Self-driving cars are still not widespread.

People are constantly on their phones and play video games, so the prediction of everyone using AR or VR is kind-of true. It just isn't immersive yet. And, unfortunately, the applications are mostly for entertainment, not for the benefit of humanity as a whole.

Computing power seems to be on schedule to eclipse the brainpower of humankind.

The AR and VR glasses are available, but they are a market "flop" thus far.

We are very close to a post-scarcity world, which Kurzweil predicted for 2020.

As was mentioned several times in the thread, a lot of these developments sound good, but at what cost? Our freedom? Having cameras and computers in everything is a privacy/freedom nightmare. Why continue down this path if we are only going to become a soul-less, freedom-less, borg collective.

Coming to the end of another year and now 2 years past the list of predictions.

It still looks pretty good, but there are obvious misses.

There are several items on the list that predict simulated people for business, teaching, socializing, but that has not happened or is not widespread.

Wireless communication is widespread but wireless technology is not yet embedded in people or clothing.

Most people still do their own driving.

Haptic interfaces are still lagging way behind.

Virtual artists (creative computers) are here but not widespread.

Household robots are not ubiquitous and reliable - except for vacuum cleaners.

What I am still dismayed about is that where the adoption of new technology is the greatest is in the realm of gaming, porn, social media...etc, things that are causing a lot of health and mental problems in people. Instead of being a positive force in society, these things have been somewhat harmful so far.

Hey Mind!

What a coincidence to check this site just after you posted this. Many of my predictions that begin the early 80s turned out to be very close to Kurzweil's, and I had no idea someone else was doing this type of thing assuming continued exponential computer power until I looked up "future brain repair" in 2004 and saw his talk at Berkeley in 2001, which is no longer at his site. I first saw him on an ABC future biology special in 1998 but had no idea who he was at the time he was briefly interviewed at the end and said: "By 2100, man and machine will have completely merged." I wondered: "How does he know anything beyond 2040 to 2060 - what I concluded in 1989 would be a "black hole like a singularity" that you couldn't see beyond.

Sadly, we were both wrong on health by 2020. I wrote in 2002 that by 2020 deaths from the major diseases apart from Alzheimer's would be slashed by 70% from their 2000 baseline. Kurzweil wrote that by 2019 we would have largely overcome the major diseases that kill 95% in the Western world. A day later, I amended what I had written and added that most cancers would be highly treatable by 2015. I hate my error here.

I actually have two half joke predictions for 2022:

First, in 2002, I wrote that there would be a nuclear war in 2022. (In 2010, I would get a flying DeLorean that would take me to 2050, and I'd be amazed at what I saw - but soon noticed how somber people were. "What's wrong? This technology is incredible", I reminded them, but one guy tuning up my flying DeLorean explained it was still hard to get over that a billion people had died 28 years earlier in The War.

Second, in 2004 I was killing time in a fast food joint and wondered when it will seem like we are in the 21st century and no longer in the 1990s. I choose 02/02/2022 at 20:22 (Tokyo time). which happens to be 4,004 days from when I thought of it and my half birthday.

You heard it here first.

 

 

Sadly, we were both wrong on health by 2020. I wrote in 2002 that by 2020 deaths from the major diseases apart from Alzheimer's would be slashed by 70% from their 2000 baseline. Kurzweil wrote that by 2019 we would have largely overcome the major diseases that kill 95% in the Western world. A day later, I amended what I had written and added that most cancers would be highly treatable by 2015. I hate my error here.

I am dismayed about these predictions that seem to be off by a few years. I have been battling for 20 years now and I thought there would be more results by now. The best interventions I have seen are Fisetin and GDF11 - just scratching the surface of the aging problem. Sadly, an enormous amount of funding is being diverted toward COVID research instead of rejuvenation and aging research. It looks like it will be up to the biohackers and maverick private businesses/entrepreneurs to cure aging.

Most of these predictions rely on intelligence. To make any of them happen requires "the humans". If above human level AI is made, we would be able to "get smarter" faster than humans can "get smarter" (humans have been upgrading this whole time, for example education and passing down things to children. But are limited in doing so).

Currently AI is making very interesting progress towards reaching human level AI. As seen on openAI.com, we already have the algorithms that generate what one could say "98% human level" text completions (extending a paragraph), image completions, music (techno, yes) completions, text2image completions, and as seen on Microsoft's NUWA's github page, video completions will soon have a higher frame rate and frame size (currently it is just sprouting). All a human brain does is predict if it is given data, and we nearly have that now, and it looks like magic when you use these AIs now, that is a really good feeling.

I have studied what would happen after AGI is made, and how easy it is to go to ASI level, and it looks very promising. It fits in with the whole evolution timeline, it's the central trunk, Cryonics is just a leaf.

Edited by Dream Big, 29 December 2021 - 06:30 PM.

Most of these predictions rely on intelligence. To make any of them happen requires "the humans". If above human level AI is made, we would be able to "get smarter" faster than humans can "get smarter" (humans have been upgrading this whole time, for example education and passing down things to children. But are limited in doing so).

 

Currently AI is making very interesting progress towards reaching human level AI. As seen on openAI.com, we already have the algorithms that generate what one could say "98% human level" text completions (extending a paragraph), image completions, music (techno, yes) completions, text2image completions, and as seen on Microsoft's NUWA's github page, video completions will soon have a higher frame rate and frame size (currently it is just sprouting). All a human brain does is predict if it is given data, and we nearly have that now, and it looks like magic when you use these AIs now, that is a really good feeling.

 

I have studied what would happen after AGI is made, and how easy it is to go to ASI level, and it looks very promising. It fits in with the whole evolution timeline, it's the central trunk, Cryonics is just a leaf.

That's all true. Exponential (AI) progress surprises people. Even if Kurzweil's predictions are off by 3 or 4 years, in the grand scheme of the evolution of the universe, that is just a "blink of an eye"....not much of a miss.

I am dismayed about these predictions that seem to be off by a few years. I have been battling for 20 years now and I thought there would be more results by now. The best interventions I have seen are Fisetin and GDF11 - just scratching the surface of the aging problem. Sadly, an enormous amount of funding is being diverted toward COVID research instead of rejuvenation and aging research. It looks like it will be up to the biohackers and maverick private businesses/entrepreneurs to cure aging.

I've wondered how much money has been spent on Covid research and how the pandemic has slowed research in all areas, but I doubt by much. We won't see much of a slow down the exponentially increasing computer power measured in MIPS per constant dollar, either.

I'm quite optimistic that there will be big advances in heart, cancer, obesity, diabetes and brain disease by the end of this decade but wish those were faster. I don't think George Church's dog research on "age reversal" has slowed down at all and he is hopeful it will be available in a decade or so. 

Edited by bluemoon, 30 December 2021 - 12:06 AM.

Since Ben Goertzel is well known in these forums, I will use his efforts as an example.

He had AI that could read medical papers way back before 2010. Ben founded OpenCog and Novamente eons ago. The field of AI is more than a half century old. I keep waiting for big breakthroughs, but there are none. When I started on the life extension journey 20 years ago, I was 30 and very healthy. Now I am feeling the affects of aging. Even though I am healthier than many of my peers, I am still getting a few grey hairs, my muscle mass is decreasing, my hair fell out, I am getting wrinkles and age spots on my skin , and many other incremental signs of aging. There is nothing to reverse these changes. There is not even anything reliable in the pipeline or near ready for human consumption/use.

I know there has been amazing progress in computing power but it has resulted in NOTHING actionable for rejuvenation. NOTHING.

Exponential progress. I know. Maybe therapies will start appearing like crazy over the next year or two. I am not so sure.

I hear ya Mind, ya. Ya, the only thing at this rate that could suddenly come jumping in with power is some ultra genius AI overlords.... That's why I have set out to focus on that area, because I know the AIs will do better than I can. They need not sleep, can think 10x faster (10 years a year), can clone adult_trained AIs, erase memories, live forever, upgrade intelligence (ability to recognize walk = whalk/ w a l k/ ruN), etc.

 

I know there has been amazing progress in computing power but it has resulted in NOTHING actionable for rejuvenation. NOTHING.

 

Exponential progress. I know. Maybe therapies will start appearing like crazy over the next year or two. I am not so sure.

 

 

I don't know about the next two years, but we should be hearing much more by 2025. Therapies are already working to end obesity, cancer treatments continue to slowly improve and by the late 2020s there will be major improvements for those with kidney and heart disease. Again, if George Church at Harvard is correct, anti-aging gene therapies will began around 2030 to 2033, and he's not the only one out there researching this. .

I don't see signs that an AI capable of symbolic manipulation and reasoning is "near", whatever that means. Besides, should we not be worried about the sanity of such a thought-generating machine? And then there's the issue of consciousness - how powerful can a non-conscious AI be?

When I started on the life extension journey 20 years ago, I was 30 and very healthy. Now I am feeling the affects of aging. Even though I am healthier than many of my peers, I am still getting a few grey hairs, my muscle mass is decreasing, my hair fell out, I am getting wrinkles and age spots on my skin , and many other incremental signs of aging. There is nothing to reverse these changes. There is not even anything reliable in the pipeline or near ready for human consumption/use.

 

I know there has been amazing progress in computing power but it has resulted in NOTHING actionable for rejuvenation. NOTHING.

 

I can totally relate to this. But I am slightly more optimistic than you regarding interventions for aging. When I started in the community 10 years ago there was nothing at all of substance that could be done. Now from older folks in the community we know things like rapamycin and GDF11 and senolytics are effective at at least increasing the quality of old age. We don't seem to have anything that can prevent youth turning into middle age yet, sadly. 

I don't think advances in computing are continuing along the exponential path Kurzweil predicted. The signs are everywhere that progress is running out of steam.