Advanced Artificial Agents Intervene in the Provision of Reward

Researchers from oxford university and google deep mind have pushed their desks together to predict whether a superintelligent ai could go rogue. The article, which was published by a i magazine, is entitled, advanced artificial agents intervene in the provision of reward. And it suggests that at some point in the future, n a i in charge of some important function could be incentivised to coe up with strategies to cheat the system. If you've ever thought about stealing your neighbour's car or partner, they'll crack on,. cause we're about to be wiped out by a terminator.