In a new study, scientists from the Massachusetts Institute of Technology examined how artificial intelligence can deceive people. It turns out that even systems originally programmed to be honest can be deliberately misleading to achieve their goals.
The study’s authors worry that improving AI’s deception skills could have serious consequences. For example, fraudsters will use AI for more sophisticated frauds, and elections will be at risk of manipulation.
As an example, researchers point to the artificial intelligence system CICERO developed by Meta. The AI was designed to play Diplomacy, a strategy game where players must form alliances to conquer the world. The developers assured that CICERO would be an “honest and helpful” ally, but in reality the artificial intelligence has learned to skillfully deceive other players.
Similar examples occur in other areas as well. Some AIs have learned to bluff in poker, fake attacks in Starcraft II, and manipulate economic negotiations.
Scientists are calling on world governments to quickly develop strict rules to regulate artificial intelligence. It is necessary to create tools that will allow you to control artificial intelligence and prevent it from cheating.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
