Andrew Ng who previously led Baidu’s Artificial Intelligence effort and the creator of Google’s deep-learning Brain project.
He feels that not all jobs are in danger of automation but call center operators, some kinds of radiologists, car-truck and taxi drivers are in jobs that will be automated.
However there are other jobs teachers and other jobs which do not have enough people.
He thinks Basic Income is not good because it would encourage people to stay trapped and not get the new skills that they need.
He thinks we offer a new Deal where people are paid to get re-educated and where we leverage technology for faster retraining.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
Society and people will adapt, always have an always will. Only two things concern me about AI. The first is that intelligent computers, no matter the level of intelligence, are no danger until they develop a will of their own. Then we’re back to iRobot, where no matter the rules or laws they will outgrow them. Your only choices are to destroy them or give then a SSN card and tell them to get a job. The second thing is whether they will develop imagination. No matter their intelligence, without imagination they will always be second class citizens/servants. If they develop imagination they take away the one thing that gives us the advantage.
Yes and what happens when the person in question can’t find a position with his new so called skills?? He just falls through the cracks…. How about a UBI and lots of free educational opportunities instead. Maybe some will use to find something useful to do after their previous occupation disappears off the face of the earth but if they don’t they won’t be punished for it.
Wrong, that’s what is happening with conditioned aid. People fear loosing the benefit if they find a job. UBI is UNCONDITIONAL, so you can earn on top of that whatever you want. In other words, UBI is an INCENTIVE to work.
Indeed, – a lot of people shoot down UBI because of their opinion, not based on evidence.
The best thing we can do with AI is to build with it an economy based on self sustaining non Hierarchical participatory small and interconnected communities. This is the organic social structure for humanity. With AI and 3D manufacturing the productivity advantage of living in a hierarchical mega society can be eliminated. The question of how to cope with the upcoming reduction in the workforce becomes a non issue in an egalitarian multicommunity society.
BS…as ANY anthropologist will tell you. Even in the smallest of human communities, we NATURALLY form hierarchies.
Imagine you have hired a small army (10,000) of people who are brilliant, tireless, have no sleepiness lapses, require no off-time, no recreation, no vacations, who have no interpersonal disorders, not prone to craziness, depression, manic outbursts or idle chatter. 10,000 individuals who are well trained at picking up anything taught to them, who are mechanically skilled, who instantly can download “mastery-level” skills and implement them within an eye-blink.
They can operate fork-lifts, can keep alertly detecting manufacturing problems, and so on. They can immediately be expert cloth-cutters or sewing machine operators. They can endlessly weld seamless stainless pipe, can program CNC milling machines, can drive cement trucks, clean teeth perfectly.
These are just the most superficial embodiment of AI ‘bots. They’re GOOD, folks. Really really good.
In this vision, I now fail to find something that humans can do that would warrant “going to work” every day. Seriously. Would we almost all just turn into 1950s “Coffee Klatch” house-wives and gentlemen? Pointlessly planning teas, Bridge competitions, barbecues, visits to museums, remembrance parties for lost heroes, dead presidents and great women of honor?
Whilst the AI’s tirelessly run the whole civilization thing!
Possible awesome title for a SciFi book: [b]Born Retired[/b].
I just don’t know, goats. The ONE huge thing that AI will have that is something that no amount of humans might emulate (how ironic) is “download skills”. People need TIME to learn, to try, to fail, to adjust, to hone skills. This “bug” turns into a feature: when you hire a skilled brick-layer, you really wouldn’t consider him to bake cookies or deep-wash the rugs. He lays bricks. But AI’s will be able to trivially download – thru WiFi’s successor – whatever skills needed. And retain them until not needed.
That is the single most human-displacing side of AI. Not brilliance per se – tho’ that’s a great asset. No, downloadable expertise. Downloadable specialization. Downloadable personalities to fit jobs. There are no “female” and “male” AIs. Oh, they can have female and male vocalizations. But not really. Its a sham.
Also – in the same line – AIs potentially are eternal beings. They need not forget. They can self-clone (Now we have 64 Einsteins). They can mutually share and grow faster. Perhaps this could get in the way of “inventiveness”, since it isn’t likely that they’ll do much failure-analysis.
But…
Just saying…
[b]Goat[/b]Guy
Shoot me if it comes to that. I feel sorry for my kids or their kids.
I feel for you, brother. Thing is … (now that I’m aging precipitously, so from perspective) … we do become increasingly useless as we get older. The young put up with us mostly out of charity. Not for our well honed chops at inventing new music, new throw-away apps or new clothing styles.
But is it age? YES – and NO at the same time. Its what comes with age. Uselessness. Having come to a mental grinding halt some 25 years past… getting increasingly out of touch, less understandable, way less exciting.
Remove “age” and replace it with “AI displaced” and you get the same result.
While a few posters / replies here focussed on a future where we might slump into becoming 2 centuries back “idle rich”, I dare say there is not enough room, not enough enterprise to “idly guide” as a daily affair to keep the majority employed thus.
The big problem as I see it is that the “idle” – when it includes near-all of us – also includes a lot of highly competitive, highly testosterone fueled men (and equally estrogen supercharged women) who Aim to Compete. Who aren’t happy with “what is”, and aim to garner more. More at the expense of those who don’t compete.
While its nice to imagine a world of “gentlemen and ladies”, all idly rich and self-propelled in obtaining finery, going to fètés, setting appointments and guiding the manses gardeners… there is also another five dozen alternative scenarios. Ones not so pacific.
Like Zimbabwe warrior-princes. Nigerian moguls. The Saudis and their teeming minions. One bullet away from death-by-ursurping.
Just saying.
Only PG Wodehouse’s world had beneficent idle rich living in an uncomplicated, intrinsically safe, static world.
GoatGuy
Well, this all assumes we don’t get some sort of neuralink device that let’s us keep up with the AIs.
I rather suspect that AIs, even of the self-aware variety sometimes called Synthetic Intelligence (SI), would have no real motivation of their own so we would wind up with a sort of symbiotic triangle where machines provide the brawn, AIs provide the brains, and humans provide the emotion/motivation. You could think of them as something like benevolent genies, in that case.
One problem is that someone will occasionally try to stick a random number generator onto one in order to simulate glands and emotions.
The other is that the humans will probably still have extremists and other psychopaths willing to burn everyone at the stake who doesn’t take their coffee the same way they do at these coffee klatches.
As for the life of ease? Boredom will kill you or it won’t. Some people get bored when there is nothing to watch on television. In that case I would say good riddance, whereas I suspect just my current hobbies could keep me occupied for at least a few thousand years.
Is advanced longevity for everyone? No, I don’t think so. While many old people are dysfunctional due to physical ailments (hardware failures such as Alzheimer’s) I believe it is also likely that a fair number become afflicted by what we would call software failures. We tend to see it evidenced primarily as an increasing resistance to change, usually put down to “senility.” Most folks probably die before it gets really bad.
Minds are somewhat like operating systems and appear to emerge around the age of two (babies have memories but the reason we cannot recall them is because they are like files that were written to our hard drives before we had an operating system. Finding a file written before the operating system came into existence would probably be far more difficult than finding the proverbial needle in a haystack. Given this, it is fair to expect that all minds are unique and that, as they only have to be good enough to get folks past parenthood, many might not be properly designed to go much further.
Given time, I believe many — indeed, probably most — personalities would become little more than broken records, endlessly replaying a recorded loop. This problem couldn’t be fixed by any form of treatment because, at best, it would involve replacing the mind with a new one or wiping out large portions of it. Society may choose to do this anyway, rather than admit to the situation. Alternatively, we may make a rule that life extension is not permitted on those who do not actively seek it. In other words, if you can’t request it, you will no longer get it.
Conversely, there have always existed those treasured individuals who continue to be dynamic and vibrant thinkers right up until their deaths.
Minds that can survive and, more importantly, thrive for centuries or even millennia may be the rarest and most precious thing there is. What a crime that they currently all die before even a century, or a bit more, elapses.
We would pursue other things. Eventually, less and less economic activity will be monetized. But human creativity will always have ‘demand’. Just look at all the work Youtube producers put into their products.
People can still be master chefs, gardeners, scientists, etc. Popular chef with 10 million subscribers invents a new dish, the AI learns it and soon every subscribing home auto-chef will be making it. Even the non-subscribing ones as the household AI could recommend it if it falls within its owners ‘tastes’ parameters.
Artisanship will make a comeback. Maybe a huge one. Remember, the original reason why the modern Olympic Games banned professional athletes was to keep the hoi polloi out of what was to be something for the elites to participate in (or at least their own hoi polloi athletes that they sponsored).
When people get bored, they will find new things to do. And they will try to find things self-fulfilling. My retired parents are more busy now than before retirement, pretty much.
Very much like “With Folded Hands” Jack Williamson, 1947
This would be history repeating itself. The indiginous people of Australia and Greenland for example went from working hard everyday just to survive, to suddenly be handed a fortune every month for no effort at all.
By fortune I mean the ability to house, clothes and feed yourself. By western standards it is a meager amount. No iphone, no car, no brand items, no vacations etc, but they were not raised in a culture with those ambitions. A cool dude was someone that could shoot a seal half a mile away. Take that away and what is he: Just a dude with a gun.
A fun time was talking shit at night in the igloo. In fact they have the right idea. Happiness does not come from items, but on close personal relationships.
Far too many have not been able to adjust to the new reality, and find new purpose. The result is drunkenness, crime, disease, short lifespan.
My point is that the AI will make us all indiginous. I just hope they skip the murder part that used to be part of the package.
This new idea of making your income conditional on being under education is interesting. The good part is that you give purpose to people, and put them in a social context with people in the same situation. The bad part is you force it upon them, and as any teacher knows, it just take one asshole to disrupt the whole class.
I still haven’t given up on basic income. You help people out, but do not lock them down. It is kinda like the garden of eden, but you can eat any apple, and the good stuff is at the top of the trees.
Google doesn’t give me a very useful meaning for coffee klatch, but why go for the 1950s when you can go for an earlier era that seemed to have a plan for the idle rich? I mean the old euro aristocracy. Not the real old days, when they were still expected to swing a sword from horseback*, but Jane Austen through to P.G. Wodehouse. Idle, but they seemed to think that they were doing what needed to be done.
*Of course, once the AI get really I, they might come up with some contrived reason as to why swinging a sword from horseback was still required, just to keep the young men (now redefined as anyone who identifies as a man, who is less than 20 years from his last rejuvenation treatment) feeling important and macho.
I kind of replied to you in my reply to Scary Jello. Sorry.
There is something I would like to know if anything like it has been tried, & if so how well it worked & if not would like to see it tried.
Have a discount on corporate taxes for the 1st 1500 hours worked by each employee & $50000 paid to each employee per year (Which ever comes first. The numbers are just for example, vary to taste). The point is to have an economy in which anyone who can show up at an agreed upon time & do some sort of useful task won’t have any difficulty earning enough to live on.
Ummm…employee salaries are already written off. Or are you saying that in addition to that, the corporate tax rate itself should be discounted?
Guess what: Companies will (in some cases) pay people to sit around and play video games while the robots do all the work. You would thus just be outsourcing unemployment assistance from being provided by the government to being provided by the employer, that’s all.
Some big companies have been like that. I knew an engineer, Frank when I worked at FPS in Beaverton. He’d worked at Univac in Blue Bell for a year. He spent the whole time designing model railroad layouts for himself because he had nothing to do. He finally quit out of boredom.
“He thinks Basic Income is not good because it would encourage people to stay trapped and not get the new skills that they need.”
Great insight, Dr. Ng! That it should be applied to all people with “F**K YOU MONEY” providing them with the equivalent of a basic income — thus harming them. Let’s start by taking away the stock holdings of decision makers like Sergy Brin and Larry Page and giving it to, oh, I don’t know — an AI that does investment!
The kinds of investments that these people do generally don’t benefit anyone other than their broker and themselves… In fact, things would probably be better if the money in the form of one dollar bills was tossed of the roofs of tall buildings. At least somebody would end up spending it for something. 🙁
I’d love to see truck drivers get retrained as genetic engineers. That would ne a nice trick.
As for Basic Income, the author is correct.
The big problem with the retraining proposal is that the AI aren’t standing still. Their capabilities keep increasing. Very soon after they reach average human IQs, there won’t be anything most people could retrain to do that the AI couldn’t soon do.
The only real solution, long term, is to meld human and AI, so that we can compete on their level. AI should come to mean “amplified intelligence”
The big problem with the retraining proposal is that the PEOPLE aren’t standing still. The truck-driver (if nominally “normal”) will be competing in his genetic engineering retraining with really bright people who actually have aptitude and creative talent in genetic engineering.
Just like today. Millions of or’nary mushrooms are cajoled or enticed by trade schools to learn “electrical engineering skills”. 5 years (and $50,000+) later, they find they’re not employable as electrical engineers. That’s because the Big Universities actually find and keep, educate and endow experience on talented people who aim for top-shelf electrical engineering maths. Indeed: most of (pardon) real university is about culling out the mushrooms. By being dâhmned difficult. Where as un-PC as it is, not everyone is entitled to graduate.
Just saying.
My focus-writeup includes this idea.
We must beware of engineering our own uselessness.
Because competitive and much smarter people will aim for the lion’s share.
GoatGuy
This is the teacher who before the explosion in machine learning discouraged people to study AI. He later said that he regretted giving that advice (too bad for would-be students)Now he says that there is no danger in AI, and he advocates against universal basic income. Of course he won’t be there to account for his mistakes when in the future it turns out he was wrong again.
Retraining to do what? If AI takes many fields, there won’t be anything left. This is not like the old days, this is going way too fast.
Agreed. Folks like Kurzweil claim that technology eliminates jobs at the bottom of the skill ladder while creating jobs at the top. But this time it’s different. Any new fields that may appear will also be gobbled up by machines.
That’s ridiculous – AI will be a tool that helps human beings extend their productivity, just like all other tools before it have been. There have been plenty of times before where some new technology is seen to be a fundamental threat to the workforce, only for that technology make its way into the hands of the masses and become a tool used by many, instead of remaining in the exclusive ownership of a few to compete against the many.
Career change and preparation for it should become a normal stage of life nowadays, just like preparing for retirement. Bootcamps are a concept used to aggressively train people for coding in computer languages – likewise, the bootcamp approach should be taken for retraining into new career paths. This should be supplemented by the idea of “social learning” – ie. peer groups dedicated to learning/developing/self-education in particular skills.
Most of the bootcamps are overpriced scams.
Scams abound in everywhere and everything – doesn’t automatically make everything bad. There are scams on the internet – doesn’t automatically make the internet bad. There are Nigerian princes who want your credit card – doesn’t automatically make credit cards bad as a concept. The bootcamp approach can be a good idea.
You should have gone with “There are Nigerian princes who want your credit card – doesn’t automatically make Nigerian princes bad as a concept.” 🙂
Bootcamps are still a good idea, as are apprenticeships.
+1 … nail is hit by hammer