The Bar For Being Human Just Got Raised Again

https://www.robotus.org/point/technological-singularityAs artificial intelligence continues to improve through both hardware and software engineering advances it’ll get to the point where it starts to overtake low functioning human beings.  Then it’ll overtake average human beings.  At some point it’ll overtake the smartest human beings on the planet.  Ray Kurzweil called this point the technological singularity.  It’s the point where we won’t know why machine intelligence is doing what it’s doing any more than an ant would understand why the humans are having a picnic.


You can read up on GPT-3, the latest step forward in A.I. here, or here, but I’m curious about it from an educational point of view.  GPT-3 is the latest iteration of OpenAI‘s research into text prediction machine intelligence.  Version three isn’t that architecturally different from GPT-2, but it’s much, much bigger, by many orders of magnitude.  This brute force approach allows it to adapt and respond much closer to a human level of response.  It’s so good it surprises people.


What does this mean in education?  GPT-3 based online systems are going to start appearing in the next year.  These systems will take a few suggestions from a human user and create text outputs that will stress a Turing test in terms of how well they are put together and what is being said.  With sufficient training and some smart engineering around focusing inputs, GPT-3 based online systems will write an accurate, original essay on any subject.  It could be used to answer any questions in any subject or formulate text responses even in abstract areas like poetry .  It’ll also translate better than anything we’ve seen so far.  It’s GPT-3’s brute-force Swiss-Army-knife effectiveness that will see it falling into student hands sooner rather than later.  Which students?  The ones it already sounds like:


“GPT-3’s ability to dazzle with prose and poetry that sounds entirely natural, even erudite or lyrical, is less surprising. It’s a parlor trick that GPT-2 already performed, though GPT-3 is juiced with more TPU-thirsty parameters to enhance its stylistic abstractions and semantic associations. As with their great-grandmother ELIZA, both benefit from our reliance on simple heuristics for speakers’ cognitive abilities, such as artful and sonorous speech rhythms. Like the bullshitter who gets past their first interview by regurgitating impressive-sounding phrases from the memoir of the CEO, GPT-3 spins some pretty good bullshit.”



Dig up some GPT-3 output online and you’ll see it uses the fact that it has figured out human speech patterns to smoothly say very little; it’s like listening to a slick salesman  This complex machine learning formula is the perfect tool for weak students answering rote, systemic school assignments, because both those students and the school system they’re responding too are so low functioning that this rudimentary A.I. can do the job better (and in less than a second).


“As AI researcher Julian Togelius put it: “GPT-3 often performs like a clever student who hasn’t done their reading, trying to bullshit their way through an exam. Some well-known facts, some half-truths, and some straight lies, strung together in what first looks like a smooth narrative.” (Though as many have pointed out: clever students who know how to bullshit go far in this world because people don’t always scrutinize what they’re saying.)”



So, the bar for human expectation has just moved again.  If you’re operating as a teacher or student at the sharp end of human achievement, this is well beneath you, but if you like to trot out the same old material year after year, don’t bother assessing process and don’t really pay much attention to student work you do mark, this’ll fool you.  For a student looking to get something for nothing, this is a dream come true.


“GPT-3 would never kill jobs skilled developers. Instead its a wake-up call for cargo coders and developers. It’ll urge them to buckle up and upskill to ensure they’re up for solving complex computer programming problems.” (cargo coders are weak programmers who copy and paste code rather than generating it themselves – they’re like many students)



The obvious answer to this is to assess process, since a student attempting to hand in work this way would have none.  Of more interest from a pedagogically standpoint is how we should integrate this evolving technology into our learning processes.  OpenAI isn’t doing this in an evil attempt to create an entire generation of illiterate children, they’re doing it to create A.I. that assists and supports human endeavour and raises it to a higher level.


The last year I was teaching English I had my 3Us try and beat Turnitin.com.  The standard usage seemed to be an ‘aha, I caught you plagiarizing!” punitive response after a minimally reviewed writing process, all done behind a curtain.  By turning that all around  and giving students transparent access to this punitive tool, I had students come to the realization that they could beat turnitin’s plagiarism check, but it took so much work to do it effectively that it was easier and more functional to just write the damned thing yourself.  Instead of depending on tech or banning it, we used it to test limitations.


I imagine education’s first response to GPT-3 driven plagiarism tools will be to try and ban them, but as usual that’s backwards.  A.I. supported human intelligence isn’t being developed for us to do less, but to enable us to do more.  From that point of view, an A.I. supported writing process should move rubric expectations for everyone upwards.  What used to meet expectations should now fail to meet expectations.  A digitally supported writer should already be leveraging tools to mitigate grammar and spelling errors, and teachers should be teaching effective use of these tools.  Where 5-10 grammar errors in a paper might have gotten you a level three/meets expectation before, there should now be none because digital supports should be integrated, proficiency in them expected and output from them meeting raised expectations.  With that technical work supported, student writers should be focusing on developing continuity of thought, voice and style.


The same goes for A.I. supported writing as we enter the Twenty-Twenties.  We should be evolving writing processes to include A.I. editorial review, A.I. supported enhanced research and maybe even A.I. driven originality of thinking.  Can you imagine a Turing test as a part of writing process that tells a student that their writing isn’t as human as a GPT-3 piece?  That’s using A.I. to raise the bar.  Can you imagine what student writing might look like if advanced word prediction A.I.s like GPT-3 were integrated into student writing processes?  We all need to be thinking about that, now.  It’s what literacy is going to look like in the next decade.


Beyond writing you’re going to see GPT-3 driven online tools rock rote, standardized (lazy) learning.  Like your worksheets?  A student will be able to scan a worksheet and receive accurate, textually correct responses instantly, to any question, in any subject.  If you’re using the same old assignments over and over, the A.I. will find that and use previous examples to produce even more complex and relevant answers.

The irony is the teachers who struggle most with this new threshold of human expectation are also the ones who will use it to mark student work.  In those teach-like-it’s-1960 classes, A.I. written papers will be handed in by students and then marked by A.I. markers – no humans will have played a part in any of that ‘learning’.



OTHER READING


https://www.wired.co.uk/article/gpt-3-openai-examples  “the world’s most impressive AI. Humans are being given limited use – for now – to make sure things don’t go wrong”


https://lambdalabs.com/blog/demystifying-gpt-3/
A technical analysis of GPT-3:  OpenAI recently published GPT-3, the largest language model ever trained. GPT-3 has 175 billion parameters and would require 355 years and $4,600,000 to train

from Blogger https://ift.tt/39LreUD
via IFTTT

Snow Garage

The snow’s flying.  I like watching Waiting out the Winter for some inspiration at this time of year:



WAITING OUT WINTER from Andrew David Watson on Vimeo.

I did my own less cool and moody version of it here:



I’m beginning to see why I’m sore after doing a few hours in the garage.  I’m all over the place!  The Tiger has new oil and filter and has been cleaned and coated and is now living under a blanket until the weather comes back to us.  The battery’s in the basement on a smart charger.



In the process of cleaning up I noticed a missing rubber on the bottom of the seat.  Gotta figure out how to get that, the cracked rubbers on the mirrors and the rubber thing that covers the rear brake wiring.  






The Tiger’s going on fourteen years old.  Fourteen years in Canada means wild swings in temperature.  The rubbers need renewing.  There’s a winter project.

from Blogger http://ift.tt/2mITyCj
via IFTTT

Motorcycle Things: Winter ’17 Wishlist

A motorcycle wish list circa 2017:



Jon Campbell on Google+ shared updated colours on the Aerostich line of motorcycle clothing.  I’ve always loved the look of Aerostich kit.  Unfortunately, a Roadcrafter suit costs more than most of the motorcycles I’ve purchased.  


One of these days I’ll get the coin together and spring for an Aerostich one piece suit.  By all accounts it’ll be the last time I need to.  


They have lots of custom options so I should be able to find a long in the body, regular inseam that fits me properly.  With colour choices aplenty, making an original looking suit that fits is an ongoing pastime.


***


Keeping with the orange kit theme, I’m also wishing for a go with the updated Desmo RO32 transformable helmet.  Quieter, more comfortable and more spacious, it’s my go-to Desmo helmet evolved.  Short of buying one from overseas untried, I’m stuck.  If we end up in France this summer, a trip to Roof might be in the cards through.


***

With the Tiger’s winter maintenance done, I’m hoping to return focus to the Concours ZG1000 Fury streetfighter I’ve got half finished.  


On the to-do list is getting a rear light and indicators.  I’d ordered them through Amazon but the dodgy Chinese company that makes them never evidently sent it, though they charged me for it.  The Amazon marketplace seems to be increasingly filled with overseas companies that have a very slow delivery time, assuming they ship at all.


It’d be nice to get this running smoothly by the summer for some blistering solo rides where I finally get to find out what those new Michelin tires feel like.  In a perfect world I’d enjoy the summer on it, ride it to the Distinguished Gentleman’s Ride in Toronto next September where someone offers to buy it for what it cost me to make it.  I could then role that over into next winter’s project.


***

A couple of road trips this summer would be nice.  I’ve had a trip around Lake Superior in mind for a while now.  It’s about 2000kms around from Manitoulin Island and back again, and another couple of hundred kilometres and a ferry ride home.


Launching from Little Current at the north end of Manitoulin, I’d go the Ontario side first just to avoid the misery that is the border crossing into Michigan at The Sault.  After sitting at that for almost two hours last year, I’ll go backwards around Superior just to avoid it.  Doing 350km/days on average, we’d get around Superior in about six days.  If we wanted a day off, we could push for a couple of days to get a day of rest.  A day up to Manitoulin and a day back at the end means eight days on the road.


A trip down the Appalachians to see the full solar eclipse this summer is also on the short list.  Doing this one for ten days means we’d have a couple of days to explore areas on the way down and on the way back instead of making miles every day.


From just over the border in New York state all the way down to Tennessee, this is motorcycle nirvana with mile after mile of twisting mountain roads.


***

Racer5 is running their introduction to track riding again this year.  A May long weekend getting familiar with the racing dynamics of a motorcycle would be pretty wicked.  By the end of the course I’d be qualified to race.  The next step would be getting myself into the VRRA for some vintage racing.

***

I never get bored of imagining throwing a few grand down on some motorcycle racing gear.  My two pairs of Alpinestar boots have been excellent, so I’d probably base a lot of the racing gear on what they offer.  I’d read reviews of the Handroid Knox racing gloves and they sound totally next level.  An Arai helmet has always been a long term, top end motorcycle helmet wishlist item, and they have a nice Isle of Man special out this year.


***

A track-day specialist bike would also be nice to have tucked away, only to be trailered to the track for hard work.   This ’99 CBR600 F4 is well cared for and going for about three grand.   I’d strip it down to bare essentials and put a carbon single seat cowling on the back.  After wearing out the tires on it, I’d go to racing tires and continue to evolve the bike into a track specialist.


***

Guy Martin did a race in his Ford Transit van last year where he averaged well over 100mph for an extended length of time.  I wouldn’t spend much time in one the other side of 100mph, but having a van would sure be handy.  From transporting my own bikes out of the snow for a cheap ride in the winter, to taking the race bike to the track, having a bike transport system would be mega.  With taxes, a new one nicely spec-ed out is just north of fifty thousand of your finest Canadian dollars.


***


Some top shelf gear, getting race ready and having the custom Kawasaki on the road… if I came into cash in 2017, that’s what I’d be spending it on.

from Blogger http://ift.tt/2ld7dLE
via IFTTT

Texas Meandering & a Better Idea

While the ice-storm of certain doom forms outside, I’m watching Qatar qualifying and daydreaming about making a MotoGP race this year.  The only one on my continent runs in a few weeks in Texas.  This has me reviewing my Texas Ironbutt dreams.

I’d originally gone for an Ironbutt on the way down and a shorter finish up the next day.  If I could push the limits I could condensify it even further (making it more excitingly possible!).



It’s just short of a twenty-four hour ride to Texas from here.  An early wake-up Friday and I could do the Ironbutt to late Friday night (60mph avg for 17 hours 4am-9pm would do it).  The last five hundred miles after an early wake up should get me to Austin on Saturday by about 2pm… just in time for qualifying.  A good sleep Saturday night and then I’m at the race Sunday.  It wraps up about 4pm.  A good push to 10pm should put me a third of the way back, I could finish up the rest on Monday.  In theory, only two days off work!






I could fly down and back if I was loaded, but a quick look around found a flight out of Detroit (4 hour drive away) leaving Friday at 10:30am and getting in to Austin just before 3pm.  Flying out of the local airport meant layovers and a long time waiting.

I found a KTM 390 Duke to rent for the four days from Lone Star Moto Rentals.  I think i could fit riding gear in carry on luggage, so there’d be no waiting for luggage and I could be in and out of the airports quickly enough.  With the bike rental, hotels and flights I’d be looking at about $3000.

By comparison the ride down would be $1000 in hotels, $200 in gas and I wouldn’t be herded onto a plane at any point.  Call me perverse but were I to go, I’d ride down.







Having said all that, I’d rather spend a thousand bucks on Racer5’s introductory track riding program. I could buy some quality race kit that’d do for years and still come in at less than this abbreviated weekend. It’d be nice to see the MotoGP boys doing their thing again, but short of an unlimited budget it doesn’t make much sense.

Big Digital Magic

I’m really enjoying teaching English again, especially the university bound group I’ve got.  I don’t have to worry about explaining why they are there as I do in many computer-tech classes.  The students come complete with their own resilience and competitive nature.  When you’re not reduced to hand holding all the time you can get into concepts deeply and quickly.

An opening unit from the text is “Fire of the Human Spirit”.  In it we look over Mandela’s inauguration Speech, a Susan Aglukark song and a June Callwood essay amongst other media, all of it pointing at the concept of FotHS.


After a few examples and some discussion we set up a wikispace where students each found a song that they believe described FotHS.  They each made a wikipage on which they provided a link to the song, the lyrics, and a personal analysis of why this song exemplifies FotHS.

Because this class comes ready to play I tend to approach it as though I’m a participant in a hot group; I like to bring gifts to the group.  In this case I knew that I could export the content out of the wikispace relatively easily.  Since that text consisted only of the lyrics and student written analysis I thought it might be interesting to look at what we’d created from a group vocabulary usage point of view.  What words found in the lyrics of 28 songs and accompanying student analysis point to our concept of Fire of the Human Spirit?

Exporting the wiki is a one click process.  Once I had the text I had to do some magic to combine all the HTML pages into a single document.  Wikispaces also exports to text but it takes the html coding with it, which made a mess.  Google-docs didn’t seem to have the mojo I needed to combine multiple documents into a single one, but the Phantom Foxit PDF creator I had did.  Once I had a pdf with all the text from twenty eight wikipages imported together I dumped it into the text window in Wordle and voila:



Katy Perry single-handedly got ‘oh’ in there!  Looking at verb usage is interesting.  Fire of the Human Spirit seems to demand action!  The nouns are also enlightening when creating constellations of meaning around this concept.  We’re going to use this class produced conglomeration of ideas to develop thesis around the concept next week.

As an aside, several English teachers turned their noses up at what we were doing.  Apparently it’s widely believed that you can’t learn English in a digital context.  I beg to differ.  If we’re going to turn to media to teach English, I’d much rather it be personalized, self created media like this.  The students themselves were surprised at how much depth something this simple offered.  That they created it as a class seemed to produce a sense of satisfaction.

Here is a FotHS 2.0 with some common words removed to emphasize specific vocabulary: