Creating Friendly AI 1.0: The Analysis and Design of Benevolent Goal Architectures

نویسنده

  • Eliezer Yudkowsky
چکیده

The goal of the field of Artificial Intelligence is to understand intelligence and create a human-equivalent or transhuman mind. Beyond this lies another question—whether the creation of this mind will benefit the world; whether the AI will take actions that are benevolent or malevolent, safe or uncaring, helpful or hostile. Creating Friendly AI describes the design features and cognitive architecture required to produce a benevolent—“Friendly”—Artificial Intelligence. Creating Friendly AI also analyzes the ways in which AI and human psychology are likely to differ, and the ways in which those differences are subject to our design decisions. Yudkowsky, Eliezer. 2001. Creating Friendly AI 1.0: The Analysis and Design of Benevolent Goal Architectures. The Singularity Institute, San Francisco, CA, June 15. The Machine Intelligence Research Institute was previously known as the Singularity Institute.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Evaluating the effect of using biophilic design and architecture strategies on the improvement of physical and mental health indicators of patients admitted to selected hospitals from the perspective of architectures, patients and patiens’ companions

  Introduction: The revolution in hospital design is called biophilic design; because the application of biophilic design in hospitals is one of the cornerstones of the humanization of health centers and hospitals. In this regard, the aim of this study was to Evaluating the effect of using biophilic design and architecture strategies on the improvement of physical and mental health indicators ...

متن کامل

Godseed: Benevolent or Malevolent?

It is known that benign looking AI objectives may result in powerful AI drives that may pose a risk to the human society. We examine the alternative scenario of what happens when universal goals that are not human-centric are used for designing AI agents. We follow a design approach that tries to exclude malevolent motivations from AI’s, however, we see that even objectives that seem benevolent...

متن کامل

Online Scheduling of Jobs for D-benevolent instances On Identical Machines

We consider online scheduling of jobs with specic release time on m identical machines. Each job has a weight and a size; the goal is maximizing total weight of completed jobs. At release time of a job it must immediately be scheduled on a machine or it will be rejected. It is also allowed during execution of a job to preempt it; however, it will be lost and only weight of completed jobs contri...

متن کامل

The New Neutral Secondary Goal based on Ideal DMU Evaluation in‏‎‎ Cross-Efficiency

Cross-efficiency is a famous ranking method for data envelopment analysis (DEA) that deletes unrealistic weights pattern with no need to a priori information ‎related ‎to ‎weights ‎restrictions. ‎This‎ method analyzes each decision making unit (DMU) taking into account the best weights resulted from assessing other DMUs. In cross-efficiency evaluation, secondary goals such as aggressiveness, be...

متن کامل

Explaining the Role of Geometry and Imagination in Creation of Unique Architectures

A In this article the role of geometry and its interaction with imagination in creating the unique architectural will be explained. From long time ago philosophers and prominents have referred to important role that imagination and geometry have in the creation of universe. All over the universe, the emergence of geometry at beginning of creation is undeniable. Geometry with its manifestation a...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1983