Published: Wed, August 08, 2018
IT | By Lester Massey

Dota 2 Pros Play Bots, Get Wrecked

Dota 2 Pros Play Bots, Get Wrecked

Original story: Research institute OpenAI, which Elon Musk co-founded in 2015, is set to send its super-smart self-taught Dota 2 bots to The International 2018 later this month to take on a team of experienced pros. "Despite that, OpenAI five thrashed some of the most popular ex Dota 2 players, including Ben "Merlini" Wu, William "Blitz" Lee, Ioannis "Fogged" Lucas, the current pro player David "MoonMeander" Tan and play-by-play commentator Austin" Capitalist" Walsh.

This was a natural follow-up to last year's bout between humans and bots, when one of OpenAI's machines defeated a top player in a one-on-one match.

Since OpenAI's Dota 2 bot beat professional player Danylo "Dendi" Ishutin at the worldwide 2017, the AI has evolved significantly. OpenAI challenged five of the world's best pro Dota 2 players, and the AI won handily. The characters Open AI played were selected by Twitch users, giving Team Human a slight advantage.


This is not the first time an AI has managed to beat humans in any game. The list of handicaps included the ability to pick only 18 of the 115 available heroes in the pool, while Team Human was given freedom to use any hero of their choosing. OpenAI Five, a team of five neural networks, may just have shown that the makers of technology, humans, are a step behind their own creation. In the third game, human players secured a win. This resulted in the bot predicting a lower chance of winning of the match.

While OpenAI's defeat of some of Dota 2's cream of the crop players is decidedly impressive, what's definitely more noteworthy is the unprecedented advancement of artificial intelligence itself. OpenAI co-founder and CTO Greg Brockman has previously described it as - if you assume that it takes a human at least 12,000 hours to become a professional at the game, OpenAI's systems essentially play 100 human lifetimes of experience every day to train.

Ahead of the three-game battle, OpenAI in a blog post explained, "Because our training system Rapid is very general, we were able to teach OpenAI Five many complex skills since June simply by integrating new features and randomizations".

Like this: