Difference between revisions of "Soft takeoff"
From Lesswrongwiki
Line 1: | Line 1: | ||
A '''soft takeoff''' refers to an AGI that would self-assemble over a period of months, years or decades. This would be due to either the learning algorithm being too demanding for the hardware or because the AI relied on experiencing feedback from the real-world that would have to be played out in real time. | A '''soft takeoff''' refers to an AGI that would self-assemble over a period of months, years or decades. This would be due to either the learning algorithm being too demanding for the hardware or because the AI relied on experiencing feedback from the real-world that would have to be played out in real time. | ||
− | Vernor Vinge, Hans Moravec and Ray Kurzweil have all expressed the view that “soft takeoff” is preferable to a “[[hard takeoff]]” as it would be both safer and easier to engineer | + | Vernor Vinge, Hans Moravec and Ray Kurzweil have all expressed the view that “soft takeoff” is preferable to a “[[hard takeoff]]” as it would be both safer and easier to engineer. |
==External Links== | ==External Links== |
Revision as of 20:45, 13 June 2012
A soft takeoff refers to an AGI that would self-assemble over a period of months, years or decades. This would be due to either the learning algorithm being too demanding for the hardware or because the AI relied on experiencing feedback from the real-world that would have to be played out in real time.
Vernor Vinge, Hans Moravec and Ray Kurzweil have all expressed the view that “soft takeoff” is preferable to a “hard takeoff” as it would be both safer and easier to engineer.
External Links
- Can we avoid a Hard takeoff? notes from Vernor Vinge.
- Robot: Mere Machine to Transcendent Mind by Hans Moravec
- The Singularity is Near by Ray Kurzweil