Can artificially intelligent agents really be conscious?

Utility function security in artificially intelligent agents

The notion of ‘wireheading’, or direct reward centre stimulation of the brain, is a wellknown concept in neuroscience. In this paper, we examine the corresponding issue of reward (utility) function integrity in artificially intelligent machines. We survey the relevant literature and propose a number of potential solutions to ensure the integrity of our artificial assistants. Overall, we conclud...

How to be Artificially Intelligent (draft)

The Pill Really Can Be Mightier Than the Sword: A Response to Recent Commentaries

We appreciate the four commentaries that add new material and fresh perspectives to our article “The pill is mightier than the sword.” In emphasizing the role of voluntary family planning and girls’ education as achievable strategies with a potential to make the world a more peaceable place, we did not mean to oversimplify or disregard the intrinsic complexity of human conflict. On the whole, t...

Adaptive Artificially Intelligent Agents in Video Games: A Survey

In the world of electronic games, the implication of entertainment has gradually moved away from developing graphics and other frills. This is because game developers have seen the need for a solid base of artificial intelligence in their game’s agents. The developers have fallen into a pattern of only using static scripts for their agents. This causes the agents to become quite predictable and...

Can working memory be non-conscious?

Working memory (WM) is closely linked to conscious awareness: In most conceptions of WM, the inputs to WM need to be conscious. The findings of some recent studies, however, have been taken to suggest that WM can indeed operate on non-conscious inputs. Here, we argue that these findings can easily be accommodated by conventional conceptions of non-conscious perception and conscious WM. We concl...