Human-based artificial intelligence nails 2016 Kentucky Derby Superfecta

TAGs: Ladbrokes, Paddy Power, Playtech, South Africa, William Hill

An artificial intelligence platform has won the “holy grail” of gambling: the Kentucky Derby Superfecta.

Human-based artificial intelligence nails 2016 Kentucky Derby SuperfectaPicking a winning from the 20-horse field at the Kentucky Derby is hard enough as it is, but things can even get a bit more hairy when bettors are asked not only to select the winner but also the second, third and fourth horse to finish the Derby in exact order. In fact, not even the experts polled by SBNation were able to pin the top four winners.

But there’s one entity—an artificial intelligence, natch—was able to get it right.

UNU, a human-based AI, accurately predicted the first, second, third and fourth horses in the Derby, prompting its inventor Louis Rosenberg and his team to win $10,842 from a $20 bet (see the official ticket here).

It was Tech Republic journalist Hope Reese who challenged Rosenberg to use the AI to predict the winners of the Kentucky Derby. UNU makes use swarm intelligence, a unique form of AI that works by having a group of volunteers to narrow down the field of 20 horses to just four, and then order the four into a Win, Place, Show, and Fourth.

Using their smartphones or computers, the participants move around a graphical magnet around the screen, dragging it to the answer they think is correct. The group only has 60 seconds to agree on one answer that best suits them all.

Rosenberg’s team wasn’t the only one who benefited from UNU’s predictions. Reese herself also placed a $1 and won about $541 on the “most exciting two minutes in sports.”

In an interview with Newsweek, Rosenberg admitted the team was reluctant to take on the challenge, but “When I saw the horses cross the line, I knew I was witnessing a milestone in the predictive abilities of AI, as well as a harbinger to future changes in how the world views sports gambling.”

This wasn’t the first winning prediction for UNU. Early this year, the AI correctly named 11 of the 15 Oscar winners, outperforming film critics and even professional gamblers. In the future, Rosenberg hopes to move into sectors like politics and health care, where doctors can use it for more accurate diagnoses.

But in the meantime, we know where to go in time for the Belmont Stakes.


views and opinions expressed are those of the author and do not necessarily reflect those of