I wrote Norvig about friendly AI issues in June but he never answered my email. I just wrote Sebastian Thrun. So if this question doesn't make it through, maybe my email will.
The Q and A are on YouTube. Not sure if there are textual transcripts. For the class itself, there is this http://www.wonderwhy-er.com/ai-class/ .
One response (unnamed): What can be done is that you can become involved yourself. I am aware of a major open source project that will be announced shortly that will have a major ethical component (I don't use the term "Friendly AI" because SIAI's proposal is shortsighted, unwise, and unethical). Contact me via my blog at http://becominggaia.wordpress.com/papers/ I (it) can point you to a number of conferences and resources as well.
Shortsighted, unwise, and unethical?
Anyone heard of this blog before?
"Becominggaia" as a website name set off my alarms, and it only got worse when I read one of his papers. The author appears to believe that greater intelligence is necessarily equated with greater benevolence, thus limiting any need for fear of existential risk. I skimmed though, and I admit I was a bit biased from the onset by the use of Kant's philosophy, so draw your own conclusions.
[...] so draw your own conclusions.
E.g. from http://vimeo.com/33767396 (slides http://bicasociety.org/2011/materials/Waser.pptx), on occasion you prefer listening to reading.
I posted an obligatory FAI question for the Thrun/Norvig office hours. You can vote here https://www.ai-class.com/sforum/?modSeries=19a660&modTopic=45&modSubmission=5cabd7. Only top questions will be answered :)
ETA: it's now the top question in the category by a nice margin!