Superhuman AI Bots Are Surprisingly Vulnerable to Exploits

  • 📰 sciam
  • ⏱ Reading Time:
  • 42 sec. here
  • 2 min. at publisher
  • 📊 Quality Score:
  • News: 20%
  • Publisher: 63%

Technology Technology Headlines News

Technology Technology Latest News,Technology Technology Headlines

By learning exploits from adversarial AI, people could defeat a superhuman Go-playing system

is heating up. But research has revealed weaknesses in one of the most successful AI systems — a bot that plays the board game Go and can beat the world’s best human players — showing that such superiority can be fragile. The study raises questions about whether more general AI systems will suffer from vulnerabilities that could compromise their safety and reliability, and even their claim to be ‘superhuman’.

In Go, two players take turns placing black and white stones on a grid to surround and capture the other player’s stones. In 2022, researchers reported, the best open-source Go-playing AI system, which typically beats the best humans handily . Their bots found exploits that regularly beat KataGo, even though the bots were otherwise not very good — human amateurs could beat them. What’s more, humans could understand the bots’ tricks and adopt them to beat KataGo.

The second defensive strategy that Gleave’s team tried was iterative: training a version of KataGo against adversarial bots, then training attackers against the updated KataGo and so on, for nine rounds. But this didn’t result in an unbeatable version of KataGo either. Adversaries kept finding exploits, with the final one beating KataGo 81% of the time.


Thank you for your comment. Your comment will be published after being reviewed.
Please try again later.
We have summarized this news so that you can read it quickly. If you are interested in the news, you can read the full text here. Read more:

 /  🏆 300. in TECHNOLOGY

Technology Technology Latest News, Technology Technology Headlines

Similar News:You can also read news stories similar to this one that we have collected from other news sources.

Reddit blocks AI bots from crawling its websiteReddit is updating its robots.txt file to block all data scrapers it doesn’t have commercial agreements with, according to a company executive.
Source: verge - 🏆 94. / 67 Read more »