astrofinch

13 ms·

61.
by astrofinch 11y ago
Could be wrong, but I believe in most cases the conversations were with users of the SL4 mailing list, and at least one user posted a PGP signed statement to the effect that the AI was let out of the box.
62.
by astrofinch 11y ago
Or like worrying about global warming back when it would have been easier to prevent? Ng's statement is, at best, equivalent to a student who is putting off starting their semester project until finals week. Yes it seems far away, but
63.
by astrofinch 11y ago
Yes, I would like to see the logic of Sam & Elon's discussions that caused them to opt for this route. Let's have them post it on the internet and give everyone an opportunity to shoot holes in it. (I hope I'd have the
64.
by astrofinch 11y ago
"If SkyNet is a threat, it's 50 or 100 years off I think." Well we work on global warming even though the worst effects are likely decades off. I think one computer science professor put it something like this: "If expe
65.
by astrofinch 11y ago
Let's say many experts thought an alien invasion was quite likely to happen in the next 100 years (the way many experts feel about AI). At that point preparation starts sounding pretty sensible, no?
66.
by astrofinch 11y ago
"Andrew Ng (I believe) compared worrying about evil AI to worrying about overpopulation on Mars." Berkeley AI prof Stuart Russell's response goes something like: Let's say that in the same way Silicon Valley companies ar
67.
by astrofinch 11y ago
It's a small demonstration of the dangers of putting computer programs that do exactly as they are programmed in charge of making important decisions. Let's say we program an AI and get one little detail wrong and things go to hel
68.
by astrofinch 11y ago
Advanced nanotechnology is not the only possible way to achieve power: http://slatestarcodex.com/2015/04/07/no-physical-substrate-n...
69.
by astrofinch 11y ago
What I've heard is that MIRI has an explicit philosophy of concentrating on the more abstract & theoretical aspects of AI safety. The idea being that if AI safety is something that you can just tack on to a working design at the e
70.
by astrofinch 11y ago
"Security through secrecy on technology has just not worked very often." Nuclear weapons come to mind. Would we prefer that the knowledge of how to make them be more widespread?
71.
by astrofinch 11y ago
It might be useful to differentiate between technological equilibria that favor either attackers or defenders. * Two guys with pistols in a crowded bar: Attacker is favored. * Trench warfare during WWI: Defender is favored. * Nukes: Attacke
72.
by astrofinch 11y ago
Musk calls that a "radically inaccurate analogy": http://lukemuehlhauser.com/musk-and-gates-on-superintelligen... AI luminary Stuart Russell also takes on this analogy in this presentation: https://www.
73.
by astrofinch 11y ago
Yes, and another issue with regulation is that it gives a speed advantage to any team that's willing to subvert the regulations: http://www.brookings.edu/blogs/techtank/posts/2015/04/14-und...
74.
by astrofinch 11y ago
Some links about this: http://blog.givewell.org/2015/02/05/putting-the-problem-of-b... https://www.againstmalaria.com/NewsItem.aspx?newsitem=Net-us...
75.
by astrofinch 11y ago
>It would be a lot more efficient to mail an envelope with cash to a random address in a disaster area than it is to expect these organizations to make a go of it. Interesting that you mention that. Charity evaluator Givewell has evalua
76.
by astrofinch 11y ago
A priori, it seems quite plausible to me that Sub-Saharan Africans find insecticide-treated bed nets significantly helpful for guarding against malaria, and rather implausible that they find them significantly harmful. You can invent far-f
77.
by astrofinch 13y ago
If you like spending time and effort preparing healthy meals for yourself, more power to you. The MealSquares target user is a busy startup founder whose diet is currently made up of microwavable meals or something like that. You tell som
78.
by astrofinch 13y ago
We see vegans as a huge potential market, because right now being a vegan is a ton of work since you have so many restrictions on how you get vital nutrients. But in the same way being a vegan is a ton of work for you, a meal replacement b
79.
by astrofinch 13y ago
(Other MealSquares cofounder here) The reason we haven't started our crowdfunding campaign yet is because our following isn't big enough yet. Soylent hit its target within the first 24 hours because the DIY recipe already had a hu
80.
by astrofinch 13y ago
It's good advice, but at the same time somewhat paradoxical. The author is essentially saying "don't force yourself to learn things; it doesn't work; instead, find something you don't have to force yourself to do a
81.
by astrofinch 13y ago
YC wasn't such a huge deal in '08 though, were they?
82.
by astrofinch 13y ago
For the super smart, there's definitely work for them to do, but it's pretty limited stuff and you really have to like physics or math to do it. Are you sure? Who's to say the super-smart couldn't help us with other m
83.
by astrofinch 13y ago
Why should a startup or tech goliath worry about concerns such as not exploiting users when people can examine the source code? When members of the coderati can customize changes, live in private clouds, and protect only themselves? What
84.
by astrofinch 13y ago
I'd expect the "domainer" to increase his asking price if the different-TLD site seems to be doing well.
85.
by astrofinch 13y ago
Could you summarize the most valuable points discussed beyond what a typical layperson already knows about the idea of flow?
86.
by astrofinch 13y ago
It's been financial news for a while: https://www.google.com/search?q=site%3Awsj.com+bitcoin
87.
by astrofinch 13y ago
There's a lot to say about Robin Hanson's argument, and I don't think it's as clear-cut as he makes it out to be, e.g. there are critical comments in this thread: http://lesswrong.com/lw/h3q/a_rational_altruist_punch_in_the... I wish he w
88.
by astrofinch 13y ago
I agree with everything you've said, but given that Against Malaria Foundation can save a life for $2,500, does treating your co-workers well and writing quality code really compare?
89.
by astrofinch 13y ago
Here's his follow up post that responds to comments in this thread: http://lionoftheblogosphere.wordpress.com/2013/05/28/follow-...
90.
by astrofinch 14y ago
Yep, they discuss that on their mistakes/shortcomings page: http://www.givewell.org/about/shortcomings#overaggressiveand... Looks like it ended in late 2007.

More ›