212

LESSWRONG
LW

211
Personal Blog

10

London meetup: "The Friendly AI Problem"

by Roko
19th Jan 2010
1 min read
28

10

Personal Blog

10

London meetup: "The Friendly AI Problem"
4MichaelGR
0CarlShulman
0dw2
3Tiiba
3Cyan
2wedrifid
0[anonymous]
0orthonormal
2[anonymous]
3Roko
5Steve_Rayhawk
10Kaj_Sotala
0wedrifid
1Kaj_Sotala
4wedrifid
2pozorvlak
0[anonymous]
1Vladimir_Nesov
4MichaelGR
9Steve_Rayhawk
3RobinZ
3[anonymous]
0JamesAndrix
0Nick_Tarleton
0wedrifid
0Paul Crowley
1ShannonVyff
0pozorvlak
New Comment
28 comments, sorted by
top scoring
Click to highlight new comments since: Today at 10:24 AM
[-]MichaelGR16y40

Could you, or someone who's attending, please film it?

It's a low cost way to reach a much bigger audience (I'd go, but I'm on another continent).

Reply
[-]CarlShulman16y00

I believe it will be put on YouTube.

Reply
[-]dw216y00

It should appear on http://www.youtube.com/user/KoanPhilosopher within a day or two.

The same channel already contains videos of several previous UKH+ meetings

Reply
[-]Tiiba16y30

Ah, so you intend to discuss a deus est machina while hammered. This will end well.

Reply
[-]Cyan16y30

I was going to message you that the expression is actually "deus ex machina", but then I Googled "deus est machina" and found the TVTropes page (no linky!).

Reply
[-]wedrifid16y20

but then I Googled "deus est machina" and found the TVTropes page (no linky!)

Thankyou. ;)

I had already pressed Cntrl-tab after parsing 'but then I googled' but I stopped myself just in time.

Reply
[-][anonymous]16y00

As a datapoint, the last meetup of this type I went to only had a fairly moderate amount of drinking. At least while I was there. Pubs are still the least worst venue for getting lots of people together to socialise in an informal setting.

Reply
[-]orthonormal16y00

deus est machina

Was that intentional or a felicitous typo? I still prefer "god from the machine" to "a god is a machine".

Reply
[-][anonymous]16y20

I'm coming a long.

Reply
[-]Roko16y30

Great. Maybe all LW'ers should wear some distinguishing mark, like a Map stuck to their clothes?

Reply
[-]Steve_Rayhawk16y50

a Map stuck to their clothes

Then a better symbol would be a surveyor's theodolite, maybe with a map (or paper), a ruler, and a drafting compass.

I've also suggested a balance scale (decision theory) with an urn full of colored balls (Bayesian inference). Another symbol could be a locket (something to protect).

Reply
[-]Kaj_Sotala16y100

I'm kinda partial to the black flowing robe filled with intricate, silver-colored mathematical equations, myself. But that might be a bit elaborate for everyday use.

Reply
[-]wedrifid16y00

Awesome. Where do I buy one?

Reply
[-]Kaj_Sotala16y10

If you're willing to pay some extra for having it handmade, there are plenty of people who would be ready to make you one.

(And now that you brought it up, you made me semi-seriously consider actually getting one made. I have a bunch of friends and friends of friends who have cosplay as a hobby, so I might be able to get it cheaper than by going through a professional. Hmm... then I'd only need to decide on the equations I wanted on it.)

Reply
[-]wedrifid16y40

Reply
[-]pozorvlak16y20

I have a half-face mask along those lines made for me by an ex-girlfriend. Sadly, she was not a mathematician, so the specific formulae aren't of much interest. Nice to have for the occasional costumed ball, though.

I met a bloke once who had Euclid's proof of Pythagoras' theorem tattooed on his arm, and got into a drunken argument with him about whether or not he should have chosen a more elegant proof.

Reply
[-][anonymous]16y00

Reply
[-]Vladimir_Nesov16y10

Go LW T-shirts!

Reply
[-]MichaelGR16y40

They design themselves.

On the front of the t-shirt on the left side you put a paper map with the word "Map" under it.

On the right, you put a 3D landscape (mountains, etc) with "Territory" written under it.

In between those two, you put "≠" (the "isn't equal to" symbol).

Voilà!

Reply
[-]Steve_Rayhawk16y90

you put a 3D landscape (mountains, etc) with "Territory" written under it.

[...]

Voilà!

("Ceci n'est pas un territoire")

Reply
[-]RobinZ16y30

It might be particularly delicious to draw a map with a copyright trap for the illustration.

Reply
[-][anonymous]16y30

I'm really tempted to make funny/silly T-shirts along the lines of

  • The anthropic principle says that the conditional probability that you are in a universe that allows reading at the moment is 1.*

  • T-shirts with one or two boxes on to indicate how you would play with Omega.

*At least according to Jurgen Schmidhuber via wikipedia because a universe where you can't read the t-shirt is inconsistent with reading the t-shirt. Yes it is a bit weak, but it is the best Anthropic principle joke I could come up with late at night.

Reply
[-]JamesAndrix16y00

I've pondered one that said "Cryocrasinating"

Reply
[-]Nick_Tarleton16y00

Blue and Green?

Reply
[-]wedrifid16y00

I'd buy a one box T-shirt if it was done well. It'd go well with my bayes rule shirt.

Reply
[-]Paul Crowley16y00

I'll be there. I look like this.

Reply
[-]ShannonVyff16y10

I'll be attending. I've been reading "Science Fiction And Philosophy" and enjoyed the section on Ethics within AI and with creating AI. I'll be bring my youngest child, age 8 who has been reading "21st Century Kids" and really thinking about AI and her own sense of self for the first time in depth. Will be a great talk, look forward to it!

Reply
[-]pozorvlak16y00

Break a leg :-)

Reply
Moderation Log
More from Roko
View more
Curated and popular this week
28Comments

To all LessWrongers in London/the south of England: On Saturday the 23rd - this Saturday at 2pm - I will be speaking about Friendly AI at the UK transhumanist group, at

Room 416, 4th floor (via main lift), Birkbeck College, Torrington Square, London (Map)

For those who have read most of the material on the web about FAI theory, e.g. the sequences here and the material of Bostrom and Omohundro, you won't hear much new, but you will probably meet a lot of people who could do with some strong rationalists to speak to. For those who haven't read most of the material about Friendly AI, I'll try to make it a superlatively informative talk. The Facebook event has 30 confirmed guests, and the London Futurists' organizer (distinct from UK h+, the hosts), just emailed me and said:

"Looking forward to meeting you and asking you questions on Saturday! I think your going to get quite a few as there is a lot of interest from my group about your presentation - positive and negative. You might get a very skeptical geezer coming along so be prepared for some tough questions from people who know their stuff."

So the event promises to be a fairly large gathering of rationalists, futurists and transhumanists with a high probability of passionate debate about FAI, followed by a trip to the pub for further discussion and drinks.