The book *Logical Foundations of Artificial Intelligence* by Michael R. Genesereth and Nils J. Nilsson is a comprehensive resource for students of AI, covering important topics in logic that are not adequately addressed in other textbooks. However, it is not suitable for a first course in AI due to significant omissions, such as the lack of discussion on expert systems, LISP, PROLOG, computer vision, and natural language processing. The book is divided into thirteen chapters, each with exercises and suggested readings, and includes answers, a bibliography, and an index. Key chapters include: 1. **Introduction**: Provides an overview and historical context. 2. **Declarative Knowledge**: Introduces first-order predicate calculus and model theory. 3. **Inference**: Focuses on rules of inference for non-automated theorem proving. 4. **Resolution**: Standard treatment of resolution theorem proving. 5. **Resolution Strategies**: Clear and readable, covering basic ideas. 6. **Nonmonotonic Reasoning**: Good chapter but could benefit from earlier examples. 7. **Induction**: Short and concise, lacking detailed examples. 8. **Reasoning with Uncertain Beliefs**: Balances probability and uncertainty but could include more examples. 9. **Knowledge and Belief**: Clear but becomes complex towards the end. 10. **Metaknowledge and Metareasoning**: Complex due to the use of quoted strings. 11. **State and Change**: Limited to a trivial example. 12. **Planning**: Describes plans for building stacks of blocks. 13. **Intelligent-Agent Architecture**: Discusses simple robot-like agents. The reviewer suggests that the book could be improved by highlighting key ideas, providing better motivation, and addressing missing topics such as truth maintenance, belief revision, and the relationship between PROLOG and pure first-order logic. The book should also include more practical examples and a fair comparison of different AI approaches. Overall, the book is a valuable resource for a second course on the use of logic in AI, but it needs revisions to make it more comprehensive and accessible.The book *Logical Foundations of Artificial Intelligence* by Michael R. Genesereth and Nils J. Nilsson is a comprehensive resource for students of AI, covering important topics in logic that are not adequately addressed in other textbooks. However, it is not suitable for a first course in AI due to significant omissions, such as the lack of discussion on expert systems, LISP, PROLOG, computer vision, and natural language processing. The book is divided into thirteen chapters, each with exercises and suggested readings, and includes answers, a bibliography, and an index. Key chapters include: 1. **Introduction**: Provides an overview and historical context. 2. **Declarative Knowledge**: Introduces first-order predicate calculus and model theory. 3. **Inference**: Focuses on rules of inference for non-automated theorem proving. 4. **Resolution**: Standard treatment of resolution theorem proving. 5. **Resolution Strategies**: Clear and readable, covering basic ideas. 6. **Nonmonotonic Reasoning**: Good chapter but could benefit from earlier examples. 7. **Induction**: Short and concise, lacking detailed examples. 8. **Reasoning with Uncertain Beliefs**: Balances probability and uncertainty but could include more examples. 9. **Knowledge and Belief**: Clear but becomes complex towards the end. 10. **Metaknowledge and Metareasoning**: Complex due to the use of quoted strings. 11. **State and Change**: Limited to a trivial example. 12. **Planning**: Describes plans for building stacks of blocks. 13. **Intelligent-Agent Architecture**: Discusses simple robot-like agents. The reviewer suggests that the book could be improved by highlighting key ideas, providing better motivation, and addressing missing topics such as truth maintenance, belief revision, and the relationship between PROLOG and pure first-order logic. The book should also include more practical examples and a fair comparison of different AI approaches. Overall, the book is a valuable resource for a second course on the use of logic in AI, but it needs revisions to make it more comprehensive and accessible.