Mint Asia ST - - News -

We wanted fly­ing cars, in­stead we got 140 char­ac­ters” - Peter Thiel

This may seem like an odd time to ques­tion the in­ge­nu­ity of the tech­nol­ogy sec­tor.

With news of artificial in­tel­li­gence (AI), au­to­ma­tion and ma­chine-learn­ing swirling around, the tech­nol­ogy space looks like a bright spot in a dull econ­omy. Mega ven­ture funds and frothy startup val­u­a­tions seem to bear out this op­ti­mism.

Yet, fifty years ago we imag­ined a very dif­fer­ent fu­ture. Stan­ley Kubrick’s clas­sic 2001: A Space Odyssey re­leased dur­ing the hey­day of the space pro­gramme as­sumed that by the early part of this mil­len­nium, space travel would be rou­tine; ma­chines would be ter­ri­fy­ingly in­tel­li­gent; and com­put­ers and hu­mans would have in­tel­li­gent con­ver­sa­tion.

By that yard­stick, the present is de­cid­edly un­der­whelm­ing. Our phones have be­come smarter, in­ter­faces slicker, and com­mu­ni­ca­tion faster. But other predictions haven’t come to pass.

Most space pro­grammes are lim­ited to un­manned ex­pe­di­tions. Af­ter the lull of the past few decades, space travel is in the news again.

But even Elon Musk, the eter­nally con­fi­dent founder of Space X, ex­pects to send a manned ex­pe­di­tion to the moon only in 2023, the first lu­nar jour­ney by hu­mans since 1972.

Driver­less ve­hi­cles are not ready to re­place hu­mans yet. The va­garies of hu­man traf­fic are just too much for these or­dered sys­tems. Robots, which are used ex­ten­sively in man­u­fac­tur­ing and dis­tri­bu­tion, haven’t been able to adapt to rou­tine hu­man tasks.

Au­to­mated as­sis­tants such as Siri are great for one-off tasks, but it’s near im­pos­si­ble to hold a con­ver­sa­tion, es­pe­cially with a thick ac­cent.

Has the tech sec­tor fallen short or were our col­lec­tive ex­pec­ta­tions un­re­al­is­tic?

Dog walk­ing, not cars fly­ing

There are around 290 uni­corns, or un­listed star­tups val­ued at more than $1 bil­lion, across the world. In the­ory, these uni­corns rep­re­sent the best of our ideas. In­vestors seem to agree, plough­ing over $980 bil­lion into these com­pa­nies.

But an anal­y­sis of the com­pa­nies shows star­tups work­ing on truly in­no­va­tive prob­lems con­sti­tute less than one-tenth of these com­pa­nies ( see chart 1).

Most of the funds (and noise) is soaked up by star­tups en­gaged in one of two things. The first, on-de­mand and e-com­merce plat­forms, find new ways to or­der things with­out leav­ing your couch.

The other class of star­tups, so­cial me­dia and en­ter­tain­ment firms are fo­cused on de­sign­ing con­tent that keeps you in that couch for as long as pos­si­ble.

Take on-de­mand firms. Uber has im­proved the taxi-hail­ing ex­pe­ri­ence; Wework has made rent­ing of­fice space easy; and with Swiggy, it’s nice not to have to walk to the restau­rant for food.

But we are still rid­ing in the same taxi, Big ideas are be­com­ing harder to find. Since more years are be­ing in­vested into find­ing the next big idea, the av­er­age age of a new en­tre­pre­neur is shift­ing. 20-34 35-44 45-54 55-64 work­ing in the same of­fice, and eat­ing from the same restau­rant as be­fore.

The or­der-to-ful­fil­ment process has been stream­lined and the sup­ply more closely matches de­mand. But this isn’t in­no­va­tion that has fun­da­men­tally al­tered how we live, work, or play. It’s just made it in­cre­men­tally bet­ter.

The de­sire to dis­rupt any ac­tiv­ity with on-de­mand or online sub­sti­tutes can some­times veer into the ab­surd. Early last year, Softbank’s Vi­sion Fund in­vested $300 mil­lion into Wag, a startup that lets you book a dog walker. (The Vi­sion Fund’s stated goal is to “in­vest in busi­nesses and foun­da­tional plat­forms that will en­able the next age of in­no­va­tion”.)

With so­cial me­dia com­pa­nies, the in­no­va­tion ar­gu­ment is even weaker. Face­book and Google evan­ge­lized the cult of con­nec­tiv­ity. But af­ter ini­tial pro­duc­tiv­ity gains from im­proved com­mu­ni­ca­tion, watch­ing YouTube videos, shar­ing Google pho­tos or brows­ing Face­book are not growth-ac­cel­er­at­ing. On the con­trary, the fa­tigue in­duced by aim­less scrolling on Twit­ter or What­sapp are a drain on pro­duc­tiv­ity.

Build­ing click­bait

For an in­dus­try that fetishizes dis­rup­tion and in­no­va­tion, this lack of cre­ativ­ity is dis­ap­point­ing. One rea­son for this state of af­fairs could be fi­nan­cial. The com­puter sci­en­tist Jaron Lanier has ar­gued that the mon­e­ti­za­tion model driv­ing the in­ter­net has cre­ated per­verse in­cen­tives for tech firms.

From the early days of the in­ter­net, we as­sumed most con­tent and ser­vices should be free. But the de­sire to de­moc­ra­tize ac­cess to in­for­ma­tion came at a cost —the ad­ver­tis­ing-sup­ported busi­ness model. The largest tech com­pa­nies to­day—face­book and Google—rely on ad­ver­tis­ing for a ma­jor­ity of their rev­enue. Even Ama­zon has now got into the game, by sell­ing pix­els on its site to ea­ger sell­ers. In the push to gen­er­ate more ad­ver­tis­ing rev­enue, these com­pa­nies need to con­tin­u­ally find ways to keep users ad­dicted (or en­gaged) to their plat­form. And as users have be­come smarter, the al­go­rithms have be­come even more in­tri­cate. To­day, some of the largest em­ploy­ers of AI and ma­chine­learn­ing talent are so­cial me­dia plat­forms. ByteDance, the cre­ator of Tik Tok and the most valu­able startup on the planet, em­ploys le­gions of AI and ma­chine­learn­ing en­gi­neers. Tik Tok does a great job of cu­rat­ing and cus­tomiz­ing con­tent, based on what’s most likely to ap­peal to a user. But peel away the glam­our and the prod­uct is still a so­cial net­work for am­a­teur mu­sic videos. A gen­er­a­tion of our best en­gi­neers are spend­ing their pro­duc­tive out­put on build­ing be­havioural nudges that ma­nip­u­late

Newspapers in English

Newspapers from Malaysia

© PressReader. All rights reserved.