The jobs we'll lose to machines -- and the ones we won't | Anthony Goldbloom

Prevoditelj: Dorian Antoniazzo Recezent: Danijela Rako

So this is my niece.

Ovo je moja nećakinja.

Her name is Yahli.

Zove se Yahli.

She is nine months old.

Ima devet mjeseci.

Her mum is a doctor, and her dad is a lawyer.

Njezina mama je liječnica, a tata je odvjetnik.

By the time Yahli goes to college,

Kada se Yahli upišše na fakultet,

the jobs her parents do are going to look dramatically different.

poslovi koje njezini roditelji rade izgledat će potpuno drugačije.

In 2013, researchers at Oxford University did a study on the future of work.

2013. godine, znanstvenici sveučilišta u Oxfordu istraživali su budućnost rada.

They concluded that almost one in every two jobs have a high risk

Zaključili su da gotovo svaki drugi posao ima visok rizik

of being automated by machines.

da bude automatiziran strojevima.

Machine learning is the technology

Strojno učenje je tehnologija

that's responsible for most of this disruption.

koja je odgovorna za većinu tog remećenja.

It's the most powerful branch of artificial intelligence.

To je najmoćnija grana umjetne inteligencije.

It allows machines to learn from data

Strojevi mogu učiti iz podataka

and mimic some of the things that humans can do.

i oponaššati neke radnje svojstvene ljudima.

My company, Kaggle, operates on the cutting edge of machine learning.

Moja tvrtka Kaggle bavi se najnaprednijim vidom strojnog učenja.

We bring together hundreds of thousands of experts

Mi povezujemo stotine tisuća stručnjaka

to solve important problems for industry and academia.

radi rješavanja važnih industrijskih i akademskih problema.

This gives us a unique perspective on what machines can do,

To nam daje jedinstveni uvid u sposobnost strojeva,

what they can't do

njihove mogućnosti

and what jobs they might automate or threaten.

i poslove koje bi mogli automatizirati ili ugroziti.

Machine learning started making its way into industry in the early '90s.

Strojno učenje postalo je dio industrije početkom 90-ih godina.

It started with relatively simple tasks.

Počelo je relativno jednostavnim zadacima.

It started with things like assessing credit risk from loan applications,

Počelo je procjenjivanjem kreditnog rizika sa zahtjeva za kredit

sorting the mail by reading handwritten characters from zip codes.

i razvrstavanjem poššte čitanjem ručno napisanih pošštanskih brojeva.

Over the past few years, we have made dramatic breakthroughs.

Kroz proteklih nekoliko godina, postigli smo nevjerojatne stvari.

Machine learning is now capable of far, far more complex tasks.

Strojno učenje sada postižže daleko, daleko naprednije rezultate.

In 2012, Kaggle challenged its community

2012. godine Kaggle je pozvao zajednicu

to build an algorithm that could grade high-school essays.

da napravi algoritam koji će ocjenjivati srednjošškolske eseje.

The winning algorithms were able to match the grades

Najbolji algoritmi dali su istu ocjenu

given by human teachers.

kao i profesori.

Last year, we issued an even more difficult challenge.

Prošle smo godine zadali još jedan zahtjevniji zadatak.

Can you take images of the eye and diagnose an eye disease

Možžete li pomoću snimke oka dijagnosticirati očnu bolest

called diabetic retinopathy?

zvanu dijabetička retinopatija?

Again, the winning algorithms were able to match the diagnoses

I ponovno, najbolji algoritmi dali su istu dijagnozu

given by human ophthalmologists.

kao i oftalmolozi.

Now, given the right data, machines are going to outperform humans

Pomoću pravilnih podataka, strojevi mogu prestići ljude

at tasks like this.

u zadacima poput ovih.

A teacher might read 10,000 essays over a 40-year career.

Profesor možže pročitati 10.000 eseja kroz 40-godišnju karijeru.

An ophthalmologist might see 50,000 eyes.

Oftalmolog možže pregledati 50.000 očiju.

A machine can read millions of essays or see millions of eyes

Stroj možže pročitati milijune eseja ili pregledati milijune očiju

within minutes.

u roku od par minuta.

We have no chance of competing against machines

Jednostavno se ne možžemo natjecati protiv strojeva

on frequent, high-volume tasks.

u čestim zadacima s mnogo podataka.

But there are things we can do that machines can't do.

No, postoje stvari koje mi možžemo, a koje strojevi ne mogu.

Where machines have made very little progress

Područje gdje su strojevi vrlo malo napredovali

is in tackling novel situations.

je rješšavanje novonastalih situacija.

They can't handle things they haven't seen many times before.

Oni se ne mogu nositi sa stvarima koje nisu vidjeli puno puta u proššlosti.

The fundamental limitations of machine learning

Osnovno ograničenje strojnog učenja

is that it needs to learn from large volumes of past data.

je to šdo mora učiti iz velike količine prijašnjih podataka.

Now, humans don't.

Ljudi ne moraju.

We have the ability to connect seemingly disparate threads

Mi imamo sposobnost spojiti naizgled nepovezane niti

to solve problems we've never seen before.

i riješšiti novonastale probleme.

Percy Spencer was a physicist working on radar during World War II,

Percy Spencer bio je fizičar koji je radio na radaru tijekom 2. svjetskog rata

when he noticed the magnetron was melting his chocolate bar.

kada je primijetio da mu magnetron otapa čokoladu.

He was able to connect his understanding of electromagnetic radiation

On je povezao svoje razumijevanje elektromagnetske radijacije

with his knowledge of cooking

sa svojim znanjem o kuhanju

in order to invent -- any guesses? -- the microwave oven.

da bi na kraju izumio -- možžete pogoditi? -- mikrovalnu pećnicu.

Now, this is a particularly remarkable example of creativity.

Ovo je jedan izvanredan primjer kreativnosti.

But this sort of cross-pollination happens for each of us in small ways

No, ovakva se povezivanja, u malim omjerima, kod svakoga od nas

thousands of times per day.

događaju tisućama puta dnevno.

Machines cannot compete with us

Strojevi se ne mogu mjeriti s nama

when it comes to tackling novel situations,

u rješavanju tih novonastalih situacija,

and this puts a fundamental limit on the human tasks

i to uvelike ograničava broj poslova u kojima

that machines will automate.

strojevi mogu zamijeniti ljude.

So what does this mean for the future of work?

I ššto to onda znači za budućnost rada?

The future state of any single job lies in the answer to a single question:

Budućnost bilo kojeg posla ovisi o odgovoru na pitanje:

To what extent is that job reducible to frequent, high-volume tasks,

Do koje se mjere taj posao možže svesti na ponavljajuće zadatke s mnogo podataka,

and to what extent does it involve tackling novel situations?

a koliko uključuje rješšavanje novonastalih situacija.

On frequent, high-volume tasks, machines are getting smarter and smarter.

U čestim zadacima s mnogo podataka, strojevi postaju sve pametniji.

Today they grade essays. They diagnose certain diseases.

Danas oni ocjenjuju eseje. Dijagnosticiraju neke bolesti.

Over coming years, they're going to conduct our audits,

S godinama će biti u stanju vrššiti revizije

and they're going to read boilerplate from legal contracts.

i čitati standardne tekstove na ugovorima.

Accountants and lawyers are still needed.

Računovođe i odvjetnici jošš su potrebni.

They're going to be needed for complex tax structuring,

Oni će biti potrebni za složžene porezne sustave

for pathbreaking litigation.

i u pravnim sporovima.

But machines will shrink their ranks

Međutim, strojevi će to promijeniti

and make these jobs harder to come by.

i smanjiti dostupnost tih poslova.

Now, as mentioned,

Kao ššto sam spomenuo,

machines are not making progress on novel situations.

strojevi ne napreduju u rješšavanju novonastalih situacija.

The copy behind a marketing campaign needs to grab consumers' attention.

Marketinšška kampanja mora privući pažžnju potroššača.

It has to stand out from the crowd.

Mora se isticati.

Business strategy means finding gaps in the market,

Poslovna strategija uključuje nalažženje rupa,

things that nobody else is doing.

stvari koje nitko drugi ne radi.

It will be humans that are creating the copy behind our marketing campaigns,

Ljudi će biti ti koji će stvarati marketinšške kampanje

and it will be humans that are developing our business strategy.

i ljudi će biti ti koji će razvijati poslovne strategije.

So Yahli, whatever you decide to do,

Tako da, Yahli, ššto god odlučila raditi,

let every day bring you a new challenge.

neka ti svaki dan donese neki novi izazov.

If it does, then you will stay ahead of the machines.

Ako tako bude, uvijek ćešš biti ispred strojeva.

Thank you.

Hvala.

(Applause)

(Pljesak)