Gallup puts median AI awareness 38 points above median use across 37 countries

Walk the ladder in Gallup's median country. Of every 100 adults, 81 know what AI is and 43 have tried it, so 53% of the aware population has ever opened one of these products ^[3][1]. Singapore gets 79 of its 96 aware adults to a first use, 82% ^[5][2]. Malawi gets 3 of 12, one in four ^[6][3].

The deeper step is proportionally less unequal than that, or rather, less unequal than the headline percentages suggest. Singapore's 46% daily users are 58% of everyone there who has ever tried AI, and Malawi's 1% daily is a third of its 3% ^[8][9]. In points of adult population the two daily-use rates sit 45 apart ^[11].

Gallup put the impact and trust questions only to adults who are aware of AI ^[18]. The 36% of Americans who expect AI to mostly help their country, third from the bottom of the 37 behind Bangladesh at 34% and Egypt at 35%, is a share of the aware rather than of all adults ^[10].

Expectation and use are not moving together. Nigeria's 68% expecting help sits against 20% who have ever used AI, a gap of 48 points, and Vietnam's 91% sits against 43%, also 48 points ^{[11][12][5][6]}.

Trust is the smaller number nearly everywhere. A majority trust AI completely or a lot for accurate information in four countries: Israel at 67%, Vietnam at 59%, China at 58% and Nigeria at 52% ^[14]. In 35 of the 37, more people expect AI to help their country than trust what it tells them ^[15]. Inside the United States, trust runs 8% among those who have never used AI, 15% among less-than-weekly users, 26% among weekly users and 45% among daily users, a factor of 5.6 from one end to the other ^[16][7]. Gallup says the survey cannot establish whether trust leads people to use AI more, whether experience builds trust, or both ^[17].

In my view the expectation numbers carry little forecasting information for anyone sizing a market. The countries at the top of that ranking are the same ones carrying 48-point gaps between what people expect and what they have used ^[11][12]. The counter-case is serious. If expectation runs ahead of use, the 57% median who have never touched AI are simply the next cohort, and the gap closes from below as price and connectivity allow ^[3].

What would settle it is the rest of the fieldwork. Gallup has completed 37 of 140 countries, 26% of the planned coverage ^[2][10], and if the countries still to come report ever-used rates close to their awareness rates, the case that the bottleneck sits between knowing and trying weakens. So would daily use climbing off Malawi's 1% while awareness stays at 12% ^[6]. In the median market as it stands, money spent on awareness is chasing the 19 points of adults who do not know what AI is, against 38 points between awareness and first use ^[12][4].