This report covers two main AI areas: natural language processing and expert systems. The discussion of each area includes an assessment of the state-of-the-art, an enumeration of problems areas and opportunities, recommendations for the next 5-10 years, and an assessment of the resources required to carry them out.

Probabilistic rules and their variants have recently supported several successful applications of expert systems, in spite of the difficulty of committing informants to particular conditional probabilities or ";certainty factors"; and in spite of the experimentally observed insensitivity of system performance to perturbations of the chosen values. Here we survey recent developments concerning reasoned assumptions which offer hope for avoiding the practical elusiveness of probabilistic rules while retaining theoretical power, for basing systems on the information unhesitatingly gained from expert informants, and reconstructing the entailed degrees of belief later.

The primary goal of the Artificial Intelligence Laboratory is to understand how computers can be made to exhibit intelligence. Two corollary goals are to make computers more useful and to understand certain aspects of human intelligence. Current research includes work on computer robotics and vision, expert systems, learning and commonsense reasoning, natural language understanding, and computer architecture.

The Fairchild Laboratory for Artificial Intelligence Research (FLAIR) was inaugurated in October, 1980, with the purposes of introduction AI Technology into Fairchild Camera and Instrument Corporation, and of broadening the AI base of its parent company, Schlumberger Ltd. The charter of the laboratory includes basic and applied research in all AI disciplines. Currently, we have significant efforts underway in several areas of computational perception, knowledge representation and reasoning, and AI-related architectures. The current computational environment includes several large mainframes dedicated to AI research, a number of high-performance personal scientific machines, and extensive graphics capabilities.

This paper argues for the position that experimental human studies are relevant to most facets of AI research and that closer ties between AI and experimental psychology will enhance the development of booth the principles of artificial intelligence and their implementation in computers. Raising psychological assumptions from the level of ad hoc intuitions to the level of systematic empirical observation, in the long run, will improve the quality of AI research and help to integrate it with related studies in other disciplines.

Our group's work in medical decision making has led us to formulate a framework for expert system design, in particular about how the domain knowledge may be decomposed into substructures. We propose that there exist different problem-solving types, i.e., uses of knowledge, and corresponding to each is a separate substructure specializing in that type of problem-solving. This is in contrast to the currently dominant expert system paradigm which proposes a common knowledge base accessed by knowledge-free problem-solvers of various kinds. In novice, these expert structures are often incomplete, and other knowledge structures and learning processes are needed to construct and complete them.

Interactive intelligent systems often suffer from a basic conflict between their computationally intensive nature and the need for responsiveness to a user. This paper introduces the Interviewer/Reasoner model, which helps to reduce this conflict. The Interviewer's primary function is to gather data while providing an acceptable response time to the user. The Reasoner does most of the symbolic computation for the system.

We describe a program for verifying that a set of rules in an expert system comprehensively spans the knowledge of a specialized domain. The program has been devised and tested within the context of the ONCOCIN System, a rule-based consultant for clinical oncology. The stylized format of ONCOIN's rule has allowed the automatic detection of a number of common errors as the knowledge base has been developed. This capability suggests a general mechanism for correcting many problems with knowledge base completeness and consistency before they can cause performance errors.

A major strength of frame-based knowledge representation languages is their ability to provide the knowledge base designer with a concise and intuitively appealing means expression. To be effective as a knowledge base development tool, a language needs to be supported by an implementation that facilitates creating, browsing, debugging, and editing the descriptions in the knowledge base. We have focused on providing such support in a SmallTalk (Ingalls, 1978) implementation of the KL-ONE knowledge representation language (Brachman, 1978), called KloneTalk, that has been in use by several projects for over a year at Xerox PARC. In this note, we describe those features of KloneTalk's displaybased interface that have made it an effective knowledge base development tool, including the use of constraints to automatically determine descriptions of newly created data base items.