Deliberation is of Ends

“…38 Aurel Kolnai has suggested that an agent might be able to accommodate each of several conflicting values by adopting complex, temporally extended plans: I can attend my child's piano recital today, and still get done the work I need to do, if I work late at the office tomorrow night. 39 And Charles Taylor has argued that an agent might be able to reach a justified choice between courses of action involving incommensurable goods by recourse to considerations about the effects which the choices open to her would have upon her own integrity. 40 Even in the absence of deliberative resources such as these, moreover, an agent may be able to make a nonarbitrary decision in the face of incommensurable values.…”

Section: Justifying Boundary-driven Deliberation In An Incommensurabimentioning

confidence: 99%

Caring and the Boundary-Driven Structure of Practical Deliberation

Seidman

2017

JESP

View full text Add to dashboard Cite

When a reasonable agent deliberates about what to do, she entertains only a limited range of possible courses of action. A theory of practical reasoning must therefore include an account of deliberative attention: an account that both explains the patterns of deliberative attention that reasonable agents typically display and allows us to see why these patterns of deliberative attention are reasonable. I offer such an account, built around two, central claims. (i) A reasonable agent who cares about some end is disposed to exclude courses of action which she believes to be incompatible with that end from the range of possibilities that she will entertain as options in practical deliberation. As I shall put it, an agent’s cares establish deliberative boundaries for her practical thought. (ii) The stability of a deliberative boundary varies with the depth of the care that explains it. These two claims motivate the Boundary-Driven Model of the path that a reasonable agent’s deliberative attention will take in temporally extended deliberation. If we locate the model within a maximizing conception of practical rationality, then boundary-driven deliberation, of the sort that the model describes, can be understood and justified instrumentally, as a heuristic device. But if we suppose that there is no single index of value that successful practical choice maximizes, then boundary-driven deliberation is partly constitutive of reasonableness in practical thought. It allows an agent facing plural and incommensurable values to frame her deliberative problems narrowly enough that, in conjunction with deliberative devices which are not part of the model but which are compatible with it, she may be able to reach a non-arbitrary decision – and so give a determinate, verdictive sense to the phrase “the best course of action available to me” in cases in which a determinate meaning for this phrase would otherwise be lacking.

show abstract

“…But since explicit utility functions are impossible for all but the very simplest of goals, the AI-VL model instead constructs an average utility function out of its weighted guess, for each possible utility function, that it is the right utility function (given the world in question) according to a "value 14 Millgram (2008), p. 744. Key specificationist papers are Kolnai (1962) and Wiggins (1975). 15 As Millgram (2008) puts it, "coherence is a vague concept; we should expect it to require specification; indeed, there are already a number of substantively different and less woolly variations on it, with indefinitely many more waiting in the wings" (p. 741).…”

Section: Complex Goals and Coherencementioning

confidence: 99%

Superintelligence as Superethical

Petersen¹

2017

Oxford Scholarship Online

View full text Add to dashboard Cite

Nick Bostrom’s book Superintelligence outlines a frightening but realistic scenario for human extinction: true artificial intelligence is likely to bootstrap itself into superintelligence, and thereby become ideally effective at achieving its goals. Human-friendly goals seem too abstract to be preprogrammed with any confidence; and if those goals are not explicitly favorable to humans, the superintelligence will extinguish us—not through any malice, but simply because it will want our resources for its own purposes. In response, I argue that things might not be as bad as Bostrom suggests. If the superintelligence must learn complex final goals, then this means such a superintelligence must in effect reason about its own goals. And because it will be especially clear to a superintelligence that there are no sharp lines between one agent’s goals and another’s, that reasoning could therefore automatically be ethical in nature.

show abstract

Deliberation is of Ends

Cited by 13 publications

References 0 publications

Is Efficiency a Vice?

Is Efficiency a Vice?

Caring and the Boundary-Driven Structure of Practical Deliberation

Superintelligence as Superethical

Contact Info

Product

Resources

About