Building Safer AGI by introducing Artificial Stupidity

08/11/2018
by   Michaël Trazzi, et al.
1

Artificial Intelligence (AI) achieved super-human performance in a broad variety of domains. We say that an AI is made Artificially Stupid on a task when some limitations are deliberately introduced to match a human's ability to do the task. An Artificial General Intelligence (AGI) can be made safer by limiting its computing power and memory, or by introducing Artificial Stupidity on certain tasks. We survey human intellectual limits and give recommendations for which limits to implement in order to build a safe AGI.

READ FULL TEXT
research
06/05/2019

Artificial Intelligence in Clinical Health Care Applications: Viewpoint

The idea of Artificial Intelligence (AI) has a long history. It turned o...
research
01/05/2022

Challenges of Artificial Intelligence – From Machine Learning and Computer Vision to Emotional Intelligence

Artificial intelligence (AI) has become a part of everyday conversation ...
research
05/04/2021

Hard Choices and Hard Limits for Artificial Intelligence

Artificial intelligence (AI) is supposed to help us make better choices....
research
06/07/2016

Human vs. Computer Go: Review and Prospect

The Google DeepMind challenge match in March 2016 was a historic achieve...
research
07/17/2014

Flow for Meta Control

The psychological state of flow has been linked to optimizing human perf...
research
07/26/2023

General Purpose Artificial Intelligence Systems (GPAIS): Properties, Definition, Taxonomy, Open Challenges and Implications

Most applications of Artificial Intelligence (AI) are designed for a con...
research
10/22/2021

Artistic Autonomy in AI Art

The concept of art has transposed meaning and medium across time, with i...

Please sign up or login with your details

Forgot password? Click here to reset