Ray Kurzweil 2019 Predictions
en.wikipedia.org
en.wikipedia.org
This is a persistent failure of well-known futurists, probably because it's more interesting to predict rapid advances than slow ones.
Given his belief in a technological singularity, and the underlying assumption that infinite, exponential technological expansion is not only possible but inevitable, this shouldn't surprise anyone. All of his predictions are essentially religious prophecies of the coming of the Eschaton.
Also, my impression of Kurzweil was that he didn't think in S-curves, but assumed technological progress follows an unbounded exponential curve - which is why he thinks he'll become immortal and bring his father back from the dead.
A depressing thought I've had is that maybe we will reach a point where we can go no further due to biological limitations. Think about monkeys, they have primitive tools but are effectively stagnant.
It's an objective fact our progress has slowed
https://www.technologyreview.com/s/601199/tech-slowdown-thre...
A monkey can't understand advanced mathematics. What if there are certain things about physics and the universe humans will never be able to grasp because we are simply too stupid as a species?
"AI" in its current form is like a child trying to make a car through imitation
I think we'll have to CRISPR ourselves smarter before we can really go full cybernetic augmentation
I suspect it's more that futurists think in terms of our engineering efforts being poured primarily into things that advance humanity, rather than the adtech, porn, entertainment, and media innovations that our Idiocracy actually seeks to reward.
Kurzweil could have been more accurate, had humanity been more prone to species improvement than, say, ever more addictive social media apps and ever more entertaining and higher resolution Netflix marathons.
The Age of Spiritual Machines is a catchier title and more interesting book than The Age of Cell Phones That Are Mostly Screen and You Can Call A Taxi With Them. There are plenty of people out there peddling much more luke-warm futures than Kurzweil; they just don't become well-known because safe, conservative predictions aren't that interesting.
Many/most things are not nearly as simple as transistors and computation are. Nature prefers sigmoids over exponentials anyway, so it’s basically always unrealistic to predict things getting exponentially different unless you have really good priors to let you predict when that sigmoid necks over.
[0] https://blogs.sciencemag.org/pipeline/archives/2007/11/06/an...
I'm pretty sure that Kurzweil is aware of this, given that he sees ICs already as the fifth paradigm in computation:
https://en.wikipedia.org/wiki/The_Singularity_Is_Near#/media...
The problem (it seems to me) is that there basically was no further paradigm shift after ICs so far, which would have been necessary for many of Kurzweil's predictions to come true this year.
> Most business transactions or information inquiries involve dealing with a simulated person.
"Most" is strong, but virtual agents are definitely widespread and fairly heavily loathed.
> Most people own more than one PC, though the concept of what a "computer" is has changed considerably
Undoubtedly correct.
> Cables connecting computers and peripherals have almost completely disappeared.
> Rotating computer hard drives are no longer used.
Very strong trendlines, although "almost completely" and "no longer" are not accurate.
> Massively parallel neural nets and genetic algorithms are in wide use.
Depending on how you define "wide use". Applications that rely on neural nets are very heavily used, but neural nets are closer to a niche technology than a widespread algorithmic hammer.
> All students have access to computers.
In developed countries, yes.
> 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.
Scarily, this is more true than not.
... That's pretty much the lot of the 2019 predictions that came close to coming true. At best that's a 20% hit rate for predictions, although many of these hits require reinterpreting "most" as "many" and so on.
I am still dubious of the above statement.
As I elaborated, "almost completely" is not true. But in the sense that the wireless alternatives are a strong share of the market and growing, I'd count it as a partially-true prediction.
Words have meanings, if he says "most" and it is less than 50% of the total then I'd have to say it's a miss.
I mean, just look at his 2009 predictions, if we just assessed his 2009 predictions as if they were for today I'd guess he'd still struggle to get 50% hit rate without favorable "reinterpretations".
I've always had a hard time understanding the appeal of futurists and their wild claims, especially amongst a population (techies) who I continue to assume are more skeptical in general.
In his 2009 predictions, there are 8 that are pretty thoroughly not true today, most of which have to do with medicine. There's another dozen or so that I would also flag as not-true or just-barely-true today. That still leaves about half of the predictions as being true for today.
But the general theme? Yeah, that's a pretty bad miss all around.
>Undoubtedly correct.
Undoubtedly? I think is a biased claim. "Most people" meaning the majority of people in the world? If so, no.
> Average life expectancy is over 100.
> Most human workers spend the majority of their time acquiring new skills and knowledge.
Also, just as a PSA, many people do not realize most deaf people don't consider themselves disabled, nor do they want to join the hearing world with cochlear implants bc it means leaving their hearing world. They are quite happy in their own network. For talking with hearing people, they use notes written on paper or phones.
> 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.
No matter whether they consider themselves disabled or not, hearing > not hearing, and anything they could do while deaf can still be done with 100% hearing restored.
As it turns out, there were quite many people who actually did not prefer modern civilization and especially the forced nature of it, and we now look back on those attempts as hallmarks of an embarrassingly racist era.
History suggests that we should be cautious of assuming that others share the same values we do, especially if we have never experienced the world in the same way as the other person. One thing you can't do with 100% hearing restored, for example, is enjoy true silence.
I think you're imagining CIs as shiny and perfect technology, but they aren't. Check out this article below for some details. [1] They take quite a long time to learn how to use. Also, for people born deaf, their hearing isn't restored, it never existed, so it becomes an addition and an added responsibility. It's two new skills - hearing and speech therapy - they have to learn since ASL/etc is already their first language. It's really just a random cultural conditioning that we tend to think we have to be perfect or even enhanced to have a good life, be loved, successful, etc., but there's much more to life than the hearing world.
> hearing > not hearing
Actually, it turns out "not hearing == hearing." They make friends, make money, program, cook, play sports, exercise, etc. You might say they can't do this with people who aren't deaf, so they have a smaller population to choose from, but their community is huge and well-connected. they actually don't need the rest of us. I sat next to a deaf guy on a plane and he ordered coffee with cream and sugar by typing on his iphone then read the whole plane ride.
[1] https://www.thisisinsider.com/why-deaf-people-turn-down-coch...
It is unrealistic to suggest that most deaf people would refuse an operation to have their hearing restored. This is just as crass as suggesting that someone who has learned to live a full life with a prosthetic leg would turn down the (hypothetical, futuristic) opportunity to have a brand new biological leg...
[Citation needed]
That is 80s level crap right here, he didn't seem to know people yet. Fair enough, but still.
The future is nothing like this man thinks, plus he doesn't understand the power of organic tech (the brain) - or the idea that any intelligence is hampered not by speed but by the density and depth of the "graph" it has to explore to provide context for any given answer.
Just my 2c -- I'm sure I will get hammered for it sometime in the future.
As I'm sure you know, AlphaGo/Alpha0 replaced much of the "graphs" for Go/Chess already, and as we talk, an open source clone (Lc0) is doing quite well in TCEC:
https://www.chessbomb.com/arena/2018-tcec-s14pd
I would not be surprised if this is the last year a classical engine wins that tournament.
And as tensor-based computing (an other non-Turing Machiene substrates) continue to mature, I expect more and more problems to be solved by carefully designed special purpose substrate/software combination, enabling solutions to problems that are practically unsolvable with traditional algorithms running on Turing Machines.
But, since a Tensorflow network does not _require_ all the functionality of a Turing Machine, it _allows_ you to limit you to more specialized computational models. And in doing so, you can run your system several orders of magnitude faster. (And the TPU type of processors are still at an early stage, expect at least a few more innovations to widen this gap further)
This comes with a few changes: - A lot of algorithms (for instance those using recursion) are hard to transfer to a "tensor machine", making some skills obsolete. - Many/most problems are solved using machine learning techniques, that tend to favor more stats/data science/maths than many pure programmers have. - Simultaneously, the new approaches are almost equally alien to old school statisticians, as the problem definitions (in particular the kinds of patterns to look for) tend to be quite different. - Also, the way to approach problems, estimate what problems can/cannot be solved or how complex they are, are quite different. (Requiring changed intuition.) - New development is to a large extent fueled by changes in hardware (CPU->GPU->TPU->?), and I think problem solutions will become more and more about seeing possibilities across both hardware and software, instead of treating these mostly separatly.
In sum, I think the change in emphasis, skills, methodology, etc is enough to be thought of as a paradigm shift.
These predications kind of feel like that. Kurzweil was "singing higher" than where we ended up. Just enough that it's a clear overshoot, but not but so much that he's off-key.
I mean if you're going to fantasize why not bring this forward and give imaginary relief to millions? I would have predicted this easily for 2016...or even maybe 2012. Obviously any earlier would have been ridiculous.
That's why I just find it funny to see that statement. For 2019. Generalizing to some global "underclass" that anyway doesn't exist.
100% agree. I just wanted to highlight that one of the largest obstacles to ensuring that everyone's basic needs are met (and that our grandchildren will still have a planet they can easily live on) is our politics and short-nearsightedness, not necessarily our technology.
Moreover, the 0.1% have the power they do because a very large portion of the rest of us choose not to partake in politics.
But wealth continues to play a role in determining social hierarchies, and as long as it does, there will be pople that see themselves as poor.
And so far, automation has contributed to making the relative differences greater than before, not smaller.
If presented with the purchasing power of today, 50 years ago, the average person of that time would see 2019 effectively post scarcity. They would hardly be able to imagine how they would need more wealth that what we have now.
But 50 years ago, the coffee was cooked in a kettle at home, while today, people expect to buy it at Starbucks. In East Asia, most people of 50 years ago hardly ever ate meat. Today most can eat at McDonalds, but Kobe beef may still be out of reach, etc.
As long as our expectations grow faster than our income, we will feel poorer as we get richer. Today's luxery goods will be taken for granted, and new, more expensive ones will be available.
Unless social hierarchi position somehow becomes independent of wealth, some people will continue to see themselves as poor.
The technology has always been easier than the people. No surprise that has affected his "business" too.
Edit: 20 quintillion would be 20 million teraflops. Sound right?
Do all the gpu cards add up to 20 quintillion?
RK's estimates are consistently best case estimates. But even if his estimates are wrong by a couple of decades, a lot can still happen before 2050.
What the prediction was trying to gauge is how much a computer that is equivalent in computational power to a human brain would cost. I think a case could be made that the human brain's computational power is not equaled by the faster supercomputer today, at which point he was off by several orders of magnitude (in terms of price).
Of course, tensor operations do not map 1:1 vs FLOPS, even for AI problems. On the other hand, tensor oriented hardware is currently getting faster at a higher rate than traditional GPU's.
But RK's prediction did assume that we would have reached full 3d chip manufacturing processes by now, and we may not be able to have human brain levels of performance on a single die until we do, which looks like it will take 5-10 years.
When LINPACK measures FLOPS, it doesn't count how many instructions the computer executed. It uses a formula for how many floating-point operations it takes to solve an NxN linear equation and combines that with how long it took the computer to execute. So NVidia's tensor instructions actually contribute tens of operations per instruction retired.
The prediction is attempting to model when a computer will exceed human computational power, which was estimated by Ray Kurzweil at 20PFLOPS, although it's unclear to me if we was specifically thinking of LINPACK Rmax or some other metric.
A looser interpretation is that you could count the number of integer adds you can push through a second multiplied by vector width. The CPU component of this would be in the region of 1.5 GHz (AVX-512 clock-speed) * 64 (8-bit adds/512-bit vector) * 2 issue ports * 10-20 cores (I don't know where the $4,000 in 1999 dollars hits in terms of pricepoint here [1]). You're looking at the region of low single-digit TOPS here. I don't know enough about the physical architecture of GPUs to estimate their theoretical headcount here, but I'd be surprised if you could push much past the teens of TOPS on those machines either.
So, for the price point he predicted, we're off by a factor of 1,000.
[1] I'm counting physical cores here. Hyperthreads don't add more actual ALUs, they just let two independent threads fill these up with operations.
By my reading about half his projections are on target and surprisingly few are completely absurd. His biggest area of concentrated failure was probably around VR/AR/Haptics, likely a result of writing the list during the first irrational exuberance period around VR and VRML.
Most of the "correct" predictions are of the following nature, namely that they are not absolute in nature:
"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"
I marked some correct because there are examples of it existing, but the wording didn't indicate it was absolute for everyone, such as:
"People with spinal cord injuries can walk and climb steps using computer-controlled nerve stimulation and exoskeletal robotic walkers."
I marked many wrong because wording was absolute such as:
"All students have access to computers."
If the prediction would have been "Most..." or something more granular I think we could agree this is probably true.
Many of the ones that are wrong are very wrong such as:
"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."
- By 2020, there will be a new World government.
I wonder what is going to have to change in 2019 for this to be true? Are we already there?
And even on that front, Moore’s law died somewhere in the 2010s. That might have thrown these off.
Because we are confronted by speculation at every turn, in print, on video, on the net,
in conversation, we may eventually conclude that it must have value.
But it doesn’t. Because no matter how many people are speculating,
no matter how familiar their faces, how good their makeup and how well they are lit,
no matter how many weeks they appear before us in person or in columns, it remains
true that none of them knows what the future holds.
Ray Kurzweil went wrong, they're his predictions.
The first is the evolution of microprocessors under Moore's Law. Since about 2005, we've come to realize that we can't afford to cool chips anymore if we crank up the frequency. What that means is that the advancements in building silicon is much less efficient at being turned into sustained computation rates. So building the exponential growth curves predicted required (and still requires) revolutionary thinking about how to build computers, rather than just easy gains each generation.
The second thing that went wrong was what Derek Lowe refers to as the "Andy Grove Fallacy." This fallacy is that every hard problem (particularly in biological systems) can be solved by just shoveling computational power at it. In the modern day, we still can't model the neurological system of a nematode that has just 302 neural cells, let alone understanding how the entire organism works. Multiply that to the scale [1] of a human being, and we are truly ignorant of how things work except in the most general sweeping statements possible.
The final thing is, well, people tend to underestimate the difficulty of problems. Things such as bipedal motion or turning a doorknob are problems that seem utterly trivial to us human beings (because we learn them before we can remember the process of learning!) but are absolutely maddening to robots.
[1] Recall that complexity tends to grow exponentially with scale, not linearly--each pairwise, three-way, four-way, etc. interaction contributes to complexity.
This includes the economic, political, ecological, and social systems, as well as the actual technological systems.
> 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.
Doesn't seem like much of a prediction these days, but from the vocabulary of the past the iPad and smartwatch would qualify, and they weren't easy to predict at the time.
I remember telling my mum in the early 2000s that one day (I didn't realise quite how soon) all phones would just be touch screens. She didn't believe me, or thought she'd be dead before it happened. It is hard to imagine the future.
I'd say we're tracking behind on a lot of these but I don't think he's so far off (and it'd be interesting to check again in just a couple years). For instance:
> People communicate with their computers via two-way speech and gestures instead of with keyboards.
In the parlance of the past, touch screens, virtual assistants and face unlock all qualify.
> Most business transactions or information inquiries involve dealing with a simulated person.
Our bar for a "simulated person" is quite high these days, but again look at it from the perspective of the past. I quite often have chats with services of varying sophistications that try and emulate a natural conversation. I'm not fooled, but that isn't the prediction.
> Pinhead-sized cameras are everywhere.
Quite literally. I have a couple looking at me right now.
> Cables connecting computers and peripherals have almost completely disappeared.
Most of my peripherals actually are wireless. "Almost completely disappeared"? Sadly not. But we're probably talking years, and this was written twenty years ago.
> Rotating computer hard drives are no longer used. > Massively parallel neural nets and genetic algorithms are in wide use.
Speak for themselves.
> Thin, lightweight, handheld displays with very high resolutions are the preferred means for viewing documents.
This one is interesting because again if you immerse yourself in the time it was written, paper was a very common occurrence on a work desk. Now? Rarely.
> Worldwide economic growth has continued. There has not been a global economic collapse.
Not as far off as it might seem.
> "Virtual sex"—in which two people are able to have sex with each other through virtual reality
Increasingly true. Check back in 12 months.
> Three-dimensional nanotube lattices are the dominant computing substrate.
Oh.
I actually think that, all in all, and taken in the spirit of the time, these are surprisingly accurate. Many are of course way off, but I personally think they show a rare prescience.
1) he underestimated the degree to which all those fancy products of modern capitalism are created not by machines making machines but by the nimble fingers of vast numbers of very poorly paid people- as it must be, because the maintenance on the machines to do the same work would be more expensive than just paying humans to do it. But from kurzweil’s point of view as a wealthy westerner, that fact is invisible, as it’s designed to be- the products simply appear to come out of a black box which we are not supposed to look inside of.
2) Technological progress has undergone a sort of phase transition in the last 100 years or so, with the discovery for the first- and only- time of fundamental facts about nature like electricity, thermodynamics, how human biology works, and so forth- discoveries that can only be made once. An s-curve looks like an exponential, for a while.
That's not the issue. The real issue is knowing how to program it.
##########
Definitely:
##########
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.
(This isn't exclusive, but Siri and home assistants definitely count, imo)
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 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. (other than exceptional or specialized cases)
Massively parallel neural nets and genetic algorithms are in wide use.
Pinhead-sized cameras are everywhere. (pinhead may be stretching it, but I think 1999 Ray Kurzweil would count our cellphone-type cameras)
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. (less the glasses, but definitely the handheld displays)
Computers have made paper books and documents almost completely obsolete. (grumble)
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. (I'm counting the rise of MOOCs and things like Khan Academy in this. You may disagree)
Students still learn together and socialize, though this is often done remotely via computers.
All students have access to computers.
Language translating machines are of much higher quality, and are routinely used in conversations. (I suspect we can count this, but I don't know if it's exactly routine yet. I've used it, an I'm not cutting edge in this realm)
Effective language technologies (natural language processing, speech recognition, speech synthesis) exist.
Anyone can wirelessly access the internet with wearable devices such as computerized glasses, contacts, and watches.
Traditional computers and communication devices such as desktop PCs, laptops, and cell phones still exist, but most of their functions can be performed by wearable gadgets. Examples include reading books, listening to music, watching movies, playing games, and teleconferencing.
Worldwide economic growth has continued. There has not been a global economic collapse. (There has been, (2 actually), but we got better)
Household robots are ubiquitous and reliable. (Roomba? Google Home / Alexa? (though they may not be robotic)
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. (I'm throwing this here, because it's so close. Everything here is true except for the unassisted highway driving, and the 'very few transportation accidents' part - the tech is definitely here)
Most decisions made by humans involve consultation with machine intelligence. For example, a doctor may seek the advice of a digital assistant. A lawyer might utilize a virtual researcher. Or a shopper may receive recommendations from a software program that has learned his or her shopping habits.
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.
Virtual artists—creative computers capable of making their own art and music—emerge in all fields of the arts.
Computerized watches, clothing, and jewelry can monitor the wearers health continuously. They can detect many types of diseases and offer recommendations for treatment.
###################################
Probably, or in cutting edge cases:
###################################
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.
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.
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 (e.g. Retinal implants allow the blind to see and spinal implants coupled with mechanical legs allow the paralyzed to walk).
Devices that deliver sensations to the skin surface of their users (e.g. 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.
Prototype personal flying vehicles using microflaps exist. They are also primarily computer-controlled. (I don't know what Microflaps are. Most of the personal flying drone things I've seen are 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.
The basic needs of the underclass are met. (Not specified if this pertains only to the developed world or to all countries) (sigh)
Most flying weapons are bird-sized robots. Some are as small as insects.
############
Not so much:
############
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.
Most business transactions or information inquiries involve dealing with a simulated person. (it isn't majority online yet I think - https://www.digitalcommerce360.com/article/us-ecommerce-sale...)
The vast majority of business interactions occur between humans and simulated retailers, or between a human's virtual personal assistant and a simulated retailer. (samesies, but we're close!)
Three-dimensional nanotube lattices are the dominant computing substrate.
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.
Most roads now have automated driving systems—networks of monitoring and communication devices that allow computer-controlled automobiles to safely navigate.
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. (throwing this here. In my experience, most people anthropomorphize, but don't believe the assistants are that intelligent)
Average life expectancy is over 100.
1. Household robots are ubiquitous and reliable.
No. Household robots exist and are in an early stage of adoption and development, but are not very capable (single tasks) or reliable, and have yet to reach widespread adoption outside of wealthy, early adopters and enthusiasts.
2. Computers do most of the vehicle driving.
No. Some cars have driver assistance technologies, but mainly expensive luxury models. Most vehicles are driven by humans with zero computer assistance.
I think you are confusing the existance of a thing with the universal availability and adoption of that thing. Where Kurzweil says most he means it. Even back in 1999 there were examples of these things in research labs and so on, the prediction he was making was that they would have moved forward from there to becoming everyday household items, present in the council estates, trailer parks and so on of the world, as well as the San Francisco area...
2. Computers in an average modern car control steering, braking and acceleration to degrees not possible in the past.
It's clear that Kurzweil is more right than wrong if you read his predictions as trends and account for some degree of variation. Since he's talking about processes I am not sure what reading his predictions with a letter-of-the-law critical mindset buys you. The vast majority of what he talks about - has been invented - and is well on the way to becoming widespread. For me, it's the point of origin ("invention") of a technology and it's near-term projection that's most salient and Kurzweil nails it down.
How much of your furniture and clothing contain embedded computers? What computers do you have in your walls?
Depending on what "computer" means, we had some of those when he made the prediction.
How is "People experience 3-D virtual reality through glasses and contact lenses that beam images directly to their retinas (retinal display)." true?
Note also that in the 1990s we had people using HUD-type VR systems. Eg, Steven Mann's "WearComp". Wasn't Kurzweil's statement therefore true when it was made? Or, does it specifically require the ability to 'beam images directly to their retinas', in which case, which systems are you thinking of?
Similarly, consider "People communicate with their computers via two-way speech and gestures instead of with keyboards.". Mac OS 9 included two-way speech back in 2001. "Computer, what time is it?" So, what does "two-way speech" really mean? And, "different personalities"?
I don't think cellphone cameras are what is meant by "pinhead-sized". Something like http://news.cornell.edu/stories/2011/07/engineers-develop-le... is "pinhead sized", at "100th of a millimeter thick and one-half millimeter on each side". A cellphone camera is close to a pinhole camera or spy camera, which existed well before 1999. Do you think Kurzweil would also count the cellphone-type cameras of the early 2000s?
"Household robots are ubiquitous and reliable" depends very much on what a "robot" means. A thermostat can be a robot. A dishwasher can be a robot. You have to figure out what Kurzweil meant by "robot", in such a way that it wasn't true already when he made it.
Regarding computer driving, you wrote "Everything here is true", but it isn't true that "Computers do most of the vehicle driving". Computers do very little vehicle driving. And, what does it mean to take "control whenever the human drives recklessly"? The computer in my 1990s car with ABS took over braking on slippery surfaces because my braking method wasn't good enough. Does that count?
How does "Virtual artists—creative computers capable of making their own art and music—emerge in all fields of the arts." fall into the "definitely" category? That is, we had computers making art and music before 1999, so Kurzweil must mean something different than that in order to be called a prediction, yes?