Truth is ten­ta­tive, and cer­tain­ty elu­sive. Noth­ing is pinned. To know is to kill. What re­mains is keep­ing the con­ver­sa­tion open. It is the on­ly path to make sense of the last ed­it and to lis­ten hon­est­ly. Gen­uine un­der­stand­ing is not a soli­tary, ob­jec­tive method, but an in­ter­play of per­spec­tives. The par­tic­i­pants in a true con­ver­sa­tion are less the lead­ers of it than the led, al­low­ing the ex­change to take its own course and reach its own un­pre­dictable end.

To en­gage in gen­uine di­a­logue, one must be open to the “Oth­er” and “Oth­er­ness”. This re­quires an eth­i­cal stance of hu­mil­i­ty and sen­si­tiv­i­ty, where one risks one’s own prej­u­dices be­ing called in­to ques­tion. To­day, in hu­man learn­ing, the pace of dig­i­tal trans­for­ma­tion ex­ceeds our ca­pac­i­ty to make sense of every­thing.

Once, the iden­ti­ties and tal­ents of lead­er­ship team mem­bers were ham­mered out through re­flec­tive prac­tice, grit, serendip­i­tous mo­ments, men­tor feed­back, wis­dom ac­cu­mu­lat­ed from fail­ure, blun­ders, re­silience, and judge­ment. We now ask: What dis­ap­pears when ar­ti­fi­cial in­tel­li­gence (AI) does this for us? What is left for us to do?

AI is dis­rupt­ing the slow, de­mand­ing process of re­vi­sions, re­cur­sions, and re­runs. It re­moves the frus­tra­tion, the long hours, and the ex­pe­ri­ences that shape us. All of it is at risk of be­ing en­gi­neered away by vibe cod­ing tools like “Lov­able”, “Cur­sor” and “Cog­ni­tion”. Vibe cod­ing tools en­able de­vel­op­ers to col­lab­o­rate with AI agents through nat­ur­al lan­guage, shift­ing our fo­cus from low-lev­el syn­tax to high-lev­el vi­sion and in­tent.

They de­moc­ra­tise soft­ware de­vel­op­ment, ac­cel­er­ate work­flows, and re­con­fig­ure cog­ni­tive tasks among hu­man agents and AI. “Lov­able” al­lows any­one to build web­sites and Apps with just a writ­ten prompt. “Cog­ni­tion” is a key play­er pri­mar­i­ly through its flag­ship prod­uct, Devin. In­stead of writ­ing code line by line, de­vel­op­ers con­verse with Devin to de­scribe their goals in nat­ur­al lan­guage. Devin trans­lates these high-lev­el in­tents in­to ex­e­cutable code, han­dling in­tri­cate im­ple­men­ta­tion de­tails au­tonomous­ly.

Devin is ca­pa­ble of plan­ning, de­bug­ging, and ex­e­cut­ing mul­ti-step tasks. By of­fload­ing tech­ni­cal ex­e­cu­tion, de­vel­op­ers can fo­cus on cre­ative vi­sion and rapid pro­to­typ­ing. Teams quick­ly con­vert func­tion­al re­quire­ments de­scribed in nat­ur­al lan­guage in­to Min­i­mum Vi­able Prod­ucts (MVPs) for test­ing and val­i­da­tion.

Of­fload­ing mo­not­o­nous and time-con­sum­ing “boil­er­plate” cod­ing, such as gen­er­at­ing unit test tem­plates, API scaf­fold­ing, or im­ple­ment­ing stan­dard func­tion­al­i­ties frees hu­man de­vel­op­ers to fo­cus on high­er-val­ue work. This can pow­er cross-func­tion­al col­lab­o­ra­tion. Non-de­vel­op­ers, in­clud­ing busi­ness an­a­lysts, ag­ile prod­uct man­agers, prod­uct own­ers and de­sign­ers can cre­ate func­tion­al soft­ware us­ing plain lan­guage, which ac­cel­er­ates feed­back loops and in­no­va­tion across sprint teams.

“Cur­sor” is an AI-pow­ered code ed­i­tor built by the “Any­sphere” team. Cur­sor quick­ly turns ideas in­to work­ing code, al­low­ing im­me­di­ate test­ing of flows and in­ter­ac­tions. It of­fers in­no­v­a­tive, con­tex­tu­al code sug­ges­tions and com­ple­tions di­rect­ly in the ed­i­tor, re­duc­ing key strokes. It can ex­plain a func­tion, make changes in place, help fix er­rors, and in­tel­li­gent­ly ap­ply cor­rec­tions across mul­ti­ple files in a project. “Cur­sor” helps build spe­cialised in­ter­nal tools that per­fect­ly fit spe­cif­ic team needs.

AI will un­doubt­ed­ly ac­cel­er­ate learn­ing and change the way we work. But ac­cel­er­at­ed learn­ing is not de­vel­op­ment. Ac­cel­er­a­tion will in­crease the out­puts. De­vel­op­ment trans­forms our iden­ti­ty. These are not in­ter­change­able. AI may not be able to take over the ex­pe­ri­ences that build mas­tery. Dur­ing the COVID-19 pan­dem­ic, we worked un­der ex­tra­or­di­nary con­di­tions. The num­ber of meet­ings in­creased, and work­loads were crush­ing.

The dig­i­tal so­lu­tions cre­at­ed an over­load. There was more ac­tiv­i­ty, not nec­es­sar­i­ly bet­ter qual­i­ty think­ing. What was tak­en away was the time to think deeply and read crit­i­cal­ly, and the chance to build a re­flec­tive prac­tice that de­vel­ops from a reper­toire of skills and com­pe­ten­cies that emerge over time in a com­mu­ni­ty of prac­tice.

Spaces for re­flec­tion and deep work con­tract­ed dur­ing the pan­dem­ic. With AI, we risk re­peat­ing this er­ror on a larg­er scale. COVID-19 ex­pand­ed the vol­ume of meet­ings. AI ex­pands the vol­ume of con­tent. “Cur­sor”, “Cog­ni­tion”, and “Lov­able” en­able faster, more ag­ile sprints, re­duce fric­tion, and re­quire few­er cor­rec­tions. AI-gen­er­at­ed slide decks and sum­maries are mul­ti­ply­ing faster than squads can in­ter­pret and pri­ori­tise them. We are pro­duc­ing more but think­ing less.

Au­to­mat­ed knowl­edge leaves less space for re­flec­tive prac­tice and cre­ative work. The new re­al­i­ty is a mul­ti­ply­ing of noise that crowds out the con­di­tions for learn­ing, in­clud­ing still­ness, mind­ful­ness, and the white space for deep think­ing. Days will be­come crowd­ed with AI-gen­er­at­ed con­tent, but with few­er mo­ments to in­ter­pret, in­te­grate, and ques­tion.

Vibe cod­ing tools al­so shift work­space de­vel­op­men­tal op­por­tu­ni­ties and ex­po­sure. Lead­ers nur­ture dis­cern­ment, in­tu­ition, car­ing, moral rea­son­ing, and em­pa­thy among team mem­bers. These ca­pa­bil­i­ties are no­to­ri­ous­ly knot­ty to fos­ter. Em­pa­thy is de­vel­oped through re­peat­ed ex­po­sure to emo­tion­al nu­ance, di­a­logue, nav­i­gat­ing am­bi­gu­i­ty and ten­sion in­side the in­ter­stices of fric­tion points in in­ter­per­son­al re­la­tion­ships. Vibe cod­ing is re­shap­ing how we work, make choic­es, and learn.

Dr Fazal Ali com­plet­ed his Mas­ter’s in Phi­los­o­phy at the Uni­ver­si­ty of the West In­dies. He was a Com­mon­wealth Schol­ar who at­tend­ed the Uni­ver­si­ty of Cam­bridge, Hugh­es Hall, and was the provost of the

Uni­ver­si­ty of Trinidad and To­ba­go, the act­ing pres­i­dent, and chair­man of the Teach­ing Ser­vice Com­mis­sion. He is present­ly a con­sul­tant with the IDB. He can be reached at faza­lal­itsc@gmail.com