Ear­lier this month Econ­sul­tan­cy published an arti­cle about why auto­ma­ted con­tent will never replace human wri­ters. The arti­cle, ‘Auto­ma­ted con­tent is a thing – but should it be?’. The arti­cle was writ­ten by Magnus Lin­kla­ter of Bes­po­ke Digi­tal, which calls its­elf a digi­tal mar­ke­ting con­sul­tan­cy.


While we agree most­ly with Lin­kla­ter, he has made some assump­ti­ons in his post that are not ent­i­re­ly true.


We can begin with whe­re we agree. Lin­kla­ter wri­tes, “The pro­blem with all of this is obvious: an algo­rithm sim­ply can­not replace a human being when it comes to tone of voice, empa­thy and an inna­te under­stan­ding of an audience’s needs. Well, not yet.”


True. We are under no illu­si­on that any text we pro­du­ce is going to win a Nobel or Pulit­zer pri­ze. It would be nice if we did but, let us be honest, that is not going to hap­pen here any­ti­me soon. Robot jour­na­lism does have its limi­ta­ti­ons: our ‘robots’ can­not con­duct inter­views and are limi­ted to the data pool avail­ab­le to them.


In our view, Lin­kla­ter makes two mista­ken assump­ti­ons.


The first is that all con­tent needs a human touch. This is just wrong. Some texts just need facts. If, for examp­le, a finan­ci­al news ser­vice just wan­ted to list the per­for­mance of three stocks on a given day, a repor­ter could sit at their desk; look up the infor­ma­ti­on; cal­cu­la­te the chan­ges in stock pri­ce; com­pa­re tho­se chan­ges to pre­vious days, weeks, and mon­ths; and then sit and wri­te a report.


That may look some­thing like this, “The day’s best-per­for­ming stocks were all in the XXX indus­try, which has seen an average gain per stock in the past two weeks of x per cent. Tho­se gai­ning the most today were A, B, and C, which saw indi­vi­du­al increa­ses of X, Y, and Z per cent. The rest of the mar­ket, howe­ver, has seen a slight fall on average of x per cent across indi­vi­du­al stocks.”


The func­tion of that sen­tence is to inform, not enter­tain. And so it does not need to be well-writ­ten (alt­hough it would be nice if it was). It is the­re­fo­re a pas­sa­ge that can be easi­ly auto­ma­ted, espe­ci­al­ly sin­ce it deals in facts and figu­res. In fact, we have a pro­ject now pro­du­cing simi­lar texts at light­ning speed.


This brings us to the second assump­ti­on. Lin­kla­ter thinks that speed is not an issue. It is, espe­ci­al­ly in news. To wri­te the pas­sa­ge above might take a repor­ter thir­ty-to-forty-five minu­tes. Our sys­tem, the rtr tex­ten­gi­ne, can spit out mul­ti­ple reports like this PER SECOND. Here, it is not ana­ly­sis that is most important, but speed. And com­pu­ters beat humans on this every time.


The­re are other advan­ta­ges to auto­ma­ti­on. One of the­se is the per­so­na­li­sing of news. To go back to a stock mar­ket examp­le, it means that texts can be tailo­red to the reader. So if the report is being gene­ra­ted for someo­ne who works in the auto­mo­ti­ve indus­try, the sys­tem could give them news about the big play­ers in that mar­ket rather than inclu­de other non-rela­ted sec­tors such as phar­maceuti­cals and food­s­tuffs. Auto­ma­ti­on means that a grea­ter varie­ty of sub­jects can be cove­r­ed.


But what the aut­hor of the arti­cle does not see is whe­re com­pu­ters sur­pass humans. Our sys­tems are able to cal­cu­la­te data and spot ano­ma­li­es in one split-second, and pro­du­ce a report in the next. Humans can­not do that.


The sup­po­si­ti­on that machi­ne wri­ters will replace human wri­ters is a fal­se one. Sure, they will replace them in some aspects of their work but that is just the natu­re of machi­nes. But they are not going to be able to replace them com­ple­te­ly. Ins­te­ad, our machi­nes will but­tress the work alre­ady done by humans, taking over the hea­vy lif­ting that can be easi­ly auto­ma­ted. Think of it like wea­ring glasses–when you first get them, you do not stop using your eyes. Ins­te­ad, they act as a sup­port to an ever­y­day func­tion.


Auto­ma­ti­on is not a hypo­theti­cal; it is alre­ady here. One sta­tis­tic says near­ly four-out-of-five com­pa­nies have been using con­tent auto­ma­ti­on for the last two years. The Los Ange­les Times, for examp­les, auto­ma­tes its ear­t­h­qua­ke reports. This is not the future. It is now. But they will not kill indus­tries much like ven­ding machi­nes did not kill off cor­ner shops.


But whe­re should the aut­hor be hap­py? Well, this respon­se was writ­ten by a human as we do not have the capa­bi­li­ty to have our robots enga­ge in a dia­lo­gue. So, in that respect, pro­fes­sio­nal writers–your jobs are, and will remain, safe.