The first-ever winners of the BotPrize, a contest to design a gaming bot that can fool opponents into thinking it's a human player.
Posted 09.28.2012 at 11:08 am
14 Comments
BotPrize Fight
The UT^2 game bot faces off against an opponent.
Courtesy of Jacob Schrum
Two virtual gamers have convinced a panel of judges they were more human than the humans they competed with in a first-person shooter game, winning the five-year-old BotPrize and beating the Turing test of machine awareness. The game bots were video game characters controlled by artificially intelligent algorithms.
One was created by computer scientists at the University of Texas at Austin, and the other by a doctoral student from Romania. The bots faced off in a game called "Unreal Tournament 2004,” in which each player tries to eliminate its opponents. Along with normal character-killing weapons, each player also had a “judging gun,” which they used to tag characters as humans or bots. The bot that is judged as the most human-like is the winner. The bots created by the two teams both achieved a humanness rating of 52 percent — way higher than average human players’ humanness rating of 40 percent, according to the BotPrize. The two teams will share the $7,000 first prize from sponsor 2K Games.
Watch below as the UT Austin bot, UT^2, kills a human opponent.
The victory comes during the 100th anniversary year of the birth of Alan Turing, the mathematician and father of computer science, whose “Turing test” is still the definition of machine intelligence. The best measure of artificial intelligence is whether it can fool people into thinking it’s one of us, he said.
Why make machines seem more human? In video games as in real life, humans are unpredictable — we hold grudges, make illogical decisions, make mistakes and learn from them, and so on. Robots don’t do this very well, or very convincingly. The BotPrize is an effort to design algorithms that can do it better. Eventually, video games, helper robots and even training simulators will feel more real, using research like this.
14 Comments
To comment, please Login.
Popular Tags
Technology
140 years of Popular Science at your fingertips.
Innovation Challenges
Popular Science+ For iPad
Each issue has been completely reimagined for your iPad. See our amazing new vision for magazines that goes far beyond the printed page
Download Our App
Stay up to date on the latest news of the future of science and technology from your iPhone or Android phone with full articles, images and offline viewing
Follow Us On Twitter
Featuring every article from the magazine and website, plus links from around the Web. Also see our PopSci DIY feed
February 2013: How To Build A Hero
Engineers are racing to build robots that can take the place of rescuers. That story, plus a city that storms can't break and how having fun could lead to breakthrough science.
Also! A leech detective, the solution to America's train-crash problems, the world's fastest baby carriage, and more.
Popular on Popsci
Most Commented
Technology
- The Latest In Classroom Protection: A Handheld Bulletproof Whiteboard
- DARPA Can See You... From 17,500 Feet In The Air
- Smoke Screen Company Thinks It Has The Perfect Defense Against School Shooters
- How Chinese Hackers Infiltrated The New York Times
- Check Out The Winning Design For The Tiny New York Apartments Of Tomorrow
- Today On Mars: Opportunity Begins Its 10th Year Of Mars Roving
- Iran Has Successfully Launched A Monkey Into Space
- Starchitect Norman Foster To Design A 3-D Printed Lunar Base
- NASA Will Launch A 13,000-Square-Foot Solar Sail Next Year
- IBM's Warmth-Activated Gel Can Break Up Tough Bacterial Biofilms And Kill Superbugs
Most Emailed
Technology
- All Of The World's Undersea Cables In One Map
- Russian Rocket Crashes In The Pacific, Carrying U.S. Satellite
- Roam The Grand Canyon Virtually With Google Maps
- NASA Will Launch A 13,000-Square-Foot Solar Sail Next Year
- "Twitching" Material Dislodges Bacteria From Ship Hulls
- Starchitect Norman Foster To Design A 3-D Printed Lunar Base
- How Chinese Hackers Infiltrated The New York Times
- Popular Science Q&A: How NASA Selected The 2013 Class Of Astronauts
- Jackets Grown From Bacteria And 10 More Feats Of Bio-Engineering
- The 11 Coolest Ideas In Biodesign
Most Viewed
Technology
- All Of The World's Undersea Cables In One Map
- Russian Rocket Crashes In The Pacific, Carrying U.S. Satellite
- Roam The Grand Canyon Virtually With Google Maps
- NASA Will Launch A 13,000-Square-Foot Solar Sail Next Year
- "Twitching" Material Dislodges Bacteria From Ship Hulls
- Starchitect Norman Foster To Design A 3-D Printed Lunar Base
- How Chinese Hackers Infiltrated The New York Times
- Popular Science Q&A: How NASA Selected The 2013 Class Of Astronauts
- The 11 Coolest Ideas In Biodesign
- Jackets Grown From Bacteria And 10 More Feats Of Bio-Engineering
Online Content Director: Suzanne LaBarre | Email
Senior Editor: Paul Adams | Email
Associate Editor: Dan Nosowitz | Email
Contributing Writers:
Clay Dillow | Email
Rebecca Boyle | Email
Colin Lecher | Email
Emily Elert | Email
Intern:
Shaunacy Ferro | Email
Today on PopSci.com
Copyright © 2012 Popular Science
A Bonnier Corporation Company. All rights reserved. Reproduction in whole or in part without permission is prohibited.
So now we're teaching the programs to hunt us without regard to reason or self-preservation.
How fun.
If the suspicions prove out to be true, that IRAN is botnet-ing and attempting to screw with the USA major banks, I hope the USA responds with an AI and botnets of our own and respond monolithically on their butts!
GO AI!
Oh, the game seems fun too, lol!
If the average human received a score of 40% and the computers received a score of 52%, that seems to indicate that they are more human than human (Rob Zombie anyone?), and thus could be differentiated from the real humans by looking for more humanistic qualities.
@Raynre
We don't need to teach robots to be irrational and careless, robots already do anything we say without question.
Making them autonomous is the scary part, because then they can PLAN how and when to kill us mercilessly.
@VBwithMe
Easily corrected by telling the Ai not to try so hard next time.
killerT,
Hence the botnets that have been installed on many USA servers causing a denial of service packet attack on our USA banks currently. This program is merciless, correct.
news.cnet.com/8301-1009_3-57520363-83/wells-fargo-is-latest-bank-to-be-hit-by-cyberattacks/
Ah yes. Apply a multiplier across all variables in the code, 0.77 in this case? We shall call it the "Don't Try So Hard (DTSH) Coefficient"
The bot can only score higher than some gamers. It can never appear more human than a large percentage of humans.
Some expert players will often seem to others to be too perfect to be human ... and will be accused of cheating which is true in some cases. However, those legit players would score very low in this test so the bot in the article seems more human than they do. If you pitted the bot against 10 players like that, it would receive a score of 100 (assuming it does not make stupid un-human like errors).
That does not mean it is more human than humans ... just seemingly more human than that test group.
Far Out, this is my concern as well. It seems more like a publicity stunt than anything. Basing the level of "human-ness" on performance in a video game where IRL players are commonly accused of bot-hacking/cheating seems trivial. Also, being that the bots only achieved a score of 52% whilst humans averaged 40% shows the faults of judging human qualities in such a forum. Without in-depth interactions between participants, it is not a shock that the ratings would hover near 50/50.
Really, the test only shows that people have a difficult time telling the difference between players and bots in limited environs. But... after all that, the gamer in me would still love to get a chance to mop-up these AIs.
Did they show that a human like bot can be designed? Or that the judges dont have a good concept of what being human is like?
Question, how do you base humaness in and of itself? Do you gauge upon speed of movement, effectiveness of flanking, number of kills? I play COD Black Ops, and I do get booted every once and a while, simply because I attack in an ambush/run and gun style, and use an almost machine-like attacks, catagorizing the atacks into either of the styles.
The idea for the military is to turn a human into a machine. If they tested it on soldiers, the humans might have gotten a lower score. or higher depending on the criteria.
Get your facts first, then you can distort them as you please.
Mark Twain
And thanks Robot, I figured Iran had made a comeback for Stuxnet.
Iran must have bee pi$$ed when they Obama admitetd U.S. involvement.
Stux is in the wires now, and we made it, so really we let loose the hydra.
Get your facts first, then you can distort them as you please.
Mark Twain
Even if robots do become so self-aware, and wanted to becom the supreme beings, What if we had robotic parts too? If the robots would hunt us down and kill all humans, What if there were people who were Half-Human? What would the robots do?
How long did it take both parties to create their bots?
Awesome achievement. Judgement day is next on the bots primary goals when they become self aware. XD