BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Comprehensive AI Services - Adrià Garriga Alonso (University of C
 ambridge)
DTSTART:20190123T170000Z
DTEND:20190123T190000Z
UID:TALK118939@talks.cam.ac.uk
CONTACT:Adrià Garriga Alonso
DESCRIPTION:A common critique of the motivation for AI safety is that it m
 akes many assumptions that aren't proven\, or even likely to be true accor
 ding to some. The most glaring one is that a general AI (AGI) will take th
 e form of a recursively self-improving agent that tries to optimise for a 
 long-term goal.\n\n"Comprehensive AI services" (CAIS) is an answer to this
  critique by Eric Drexler from the Future of Humanity institute. He provid
 es an alternative model for an AGI\, which is that it emerges as a collect
 ion of AI-based software services of bounded scope and time to act. If thi
 s is a very likely scenario for the emergence of AGI\, the priorities of c
 urrent AI safety research should change.\n\nWe will discuss the CAIS model
  and its implications for safety research.\n\nReading list (as usual\, we 
 start reading at 5 pm\, but the discussion starts at 5:30 pm)\n\n- Rohin S
 hah's "summary of CAIS":https://www.alignmentforum.org/posts/x3fNwSe5aWZb5
 yXEG/reframing-superintelligence-comprehensive-ai-services-as\n\n- Richard
  Ngo's "summary of CAIS":https://www.alignmentforum.org/posts/HvNAmkXPTSoA
 4dvzv/comments-on-cais\n\n- Very long\, I suggest only reading a few choic
 e sections. It'd be good to write down which sections you read\, so we kno
 w which to ask you for a summary of and which to discuss together. Eric Dr
 exler's technical report: "Reframing superintelligence: Comprehensive AI s
 ervices as general intelligence":https://www.fhi.ox.ac.uk/reframing/\n
LOCATION:Cambridge University Engineering Department\, CBL Seminar room BE
 4-38
END:VEVENT
END:VCALENDAR