Eager to impress, the young soccer players darted and dribbled around the purple cones laid out on the patchy grass. Their speed, control and footwork were all being carefully evaluated \u2014 but not by a veteran scout.<\/p>\n
Not even by a human being.<\/p>\n
Instead, the Brazilian athletes were being assessed by a mobile app powered by artificial intelligence, part of a handful of new tools promising to revolutionize the way that talent is discovered in the soccer-mad nation.<\/p>\n
\u201cWe\u2019re talking about millions of unseen boys and girls,\u201d said Roger Wittmann, a German sports agent who created Cuju, one scouting app gaining ground in Brazil. \u201cThis is a big chance for them to be seen.\u201d<\/p>\n
Platforms like Cuju have quickly attracted hundreds of thousands of users in a country where playing professional soccer is a dream shared by many. The tools have also caught the attention of major soccer clubs, some of which are now using apps to recruit.<\/p>\n
A.I. scouting platforms are already common in Europe, where evaluating soccer talent has long been rooted in metrics and statistics. But in Brazil, deep economic and regional inequalities have historically made it difficult to standardize scouting practices.<\/p>\n
Instead, discovering Brazil\u2019s soccer talent has been left to the nation\u2019s storied scouts. Known as \u201colheiros\u201d in Portuguese, these veteran talent hunters often spend decades scouring amateur matches, neighborhood leagues and school tournaments, from the Amazon rainforest to Brazil\u2019s dusty outback, looking for the next big star.<\/p>\n
Yet now, powerful A.I. technology could identify emerging talent faster and better than the human eye alone. It also could reach more aspiring athletes across Brazil\u2019s vast territory, giving even those in remote reaches of the country, where few scouts might venture, a chance to be spotted.<\/p>\n
Most A.I. scouting platforms work by analyzing videos that users upload or drills they record directly into the apps. Weighing a vast array of abilities, from speed to ball control, they spit out a score and add athletes to a database. There, human agents seeking talent can find them, or the apps can pitch athletes directly to specific clubs.<\/p>\n
For now, at least part of talent recruitment remains in human hands. But progress in A.I. is prompting discussions about how much should be turned over to machines.<\/p>\n
On a crisp Sunday morning, a few dozen teenagers lined rows of faded concrete bleachers in a modest stadium in rural S\u00e3o Paulo State, clutching well-worn, muddy cleats.<\/p>\n
Many had come from towns hundreds of miles away for a tryout that could earn them a roster position on a local team in Agua\u00ed, a sleepy town of 30,000. The athletes, all 14 to 19, had been selected on the basis of their scores on an A.I.-powered mobile app. Now, they hoped to show off their skills in real life.<\/p>\n
On the sidelines, some boys ran one-minute drills as the app captured and graded their movements in real time. Later, on the field, the teenagers jostled for control of the ball as a team of trainers watched closely.<\/p>\n
Davi Barossi, 18, quickly made an impression. He dodged two defenders and plunged the ball into the corner of the net. Mr. Barossi had traveled 10 hours by car from the southern state of Santa Catarina. \u201cI\u2019m here pursuing my dream,\u201d he said, the day after watching Brazil\u2019s national team compete in the World Cup. <\/p>\n
Nathan Moraes, an 18-year-old from the Amazonian state of Par\u00e1, was having a tougher time. After a tackle gone wrong, he limped off the field, wincing in pain. \u201cEach opportunity you get, you have to give it your all,\u201d he said, massaging a cramped shin.<\/p>\n
During a pause, the players grilled each other, wolfed down chopped fruit and guzzled water. \u201cI\u2019m in second place on the app,\u201d Mr. Moraes boasted. \u201cWhat\u2019s your ranking?\u201d<\/p>\n
On a battered phone, Mr. Barossi showed off his own metrics. He worried because he is shorter than many players his age, he said. But, after doing drills through the app every day, he had climbed into the top 30 nationwide in his age group. \u201cI\u2019m always kicking the ball around and recording it,\u201d he said. <\/p>\n
While most experts agree that A.I. tools can help set precise, standardized criteria for players, they warn that technology can have blind spots.<\/p>\n
Metrics can favor taller or stronger athletes, overlooking less conventional talent. While much of Brazil is online, these apps remain less accessible to poor athletes without a decent connection or a quality cellphone camera. And users can delete or swap uploaded videos as many times as they want, notching a higher score that may not always reflect their abilities.<\/p>\n
Then there are those who believe that, no matter how advanced, artificial intelligence simply cannot top the trained eye of a professional olheiro.<\/p>\n
\u201cIt\u2019s a gift that God gives you,\u201d said Jo\u00e3o Maradona, a Brazilian scout whose work scouring Brazil\u2019s remote northeast unearthed several athletes who went on to play for Brazil\u2019s national team. \u201cNobody can teach you to see, in just 15 or 20 minutes, that raw talent that\u2019s really special.\u201d<\/p>\n
Perhaps unsurprisingly, those building A.I. scouting apps have different views. At a small start-up office in S\u00e3o Paulo on a recent afternoon, analysts working for the app Footbao sifted through videos submitted by athletes. With speedy clicks of a mouse, they scored each player across two dozen different categories.<\/p>\n
The grainy clips had been selected by artificial intelligence as the most promising out of tens of thousands uploaded to the app by players around Brazil. After the analysts vetted the videos, A.I. technology ranked the players using a formula and produced a detailed report for clubs. <\/p>\n
The A.I. was still being trained. But the aim was to eventually automate the evaluation, reducing human error and grounding the scouting process in data.<\/p>\n
\u201cWe\u2019re not trying to take the job away from a scout,\u201d said Nick Rappolt, the start-up\u2019s chief operating officer. \u201cWe make scouting more efficient, more economical.\u201d<\/p>\n