Speaking at the discussion, Musk said there was a role for governments to play "when the public safety is at risk" in regards to AI.
Category
🗞
NewsTranscript
00:00 I would tell people, we should really be concerned about AI.
00:01 They'd be like, what are you talking about?
00:02 I've never really had any experience with AI.
00:06 Since I was immersed in technology,
00:09 I have been immersed in technology for a long time,
00:11 I could see it coming.
00:12 (upbeat music)
00:14 Obviously, things like chat GPT were quite remarkable.
00:22 Now I saw GPT-1, GPT-2, GPT-3, GPT-4,
00:26 the whole sort of lead up to that.
00:27 So it was easy for me to kind of see where it's going.
00:32 If you just sort of extrapolate the points on a curve
00:34 and assume that trend will continue,
00:36 then we will have profound artificial intelligence
00:39 and obviously at a level
00:40 that far exceeds human intelligence.
00:42 So, I'm glad to see at this point
00:47 that people are taking safety seriously.
00:51 And I'd like to say thank you
00:53 for holding this AI safety conference.
00:55 I think actually it will go down in history
00:56 as being very important.
00:58 I do think overall that the potential is there
01:00 for artificial intelligence, AI,
01:02 to have most likely a positive effect
01:06 and to create a future of abundance
01:11 where there is no scarcity of goods and services.
01:14 But it is somewhat of the magic genie problem,
01:18 where if you have a magic genie
01:19 that can grant all the wishes,
01:22 usually those stories don't end well.
01:25 (laughing)
01:26 Be careful what you wish for, including wishes.
01:29 - What are the types of things
01:30 governments like ours should be doing
01:31 to manage and mitigate against those risks?
01:34 - Well, I generally think that it is good
01:37 for government to play a role
01:38 when the public safety is at risk.
01:41 So, really for the vast majority of software,
01:45 the public safety is not at risk.
01:46 I mean, if the app crashes on your phone, your laptop,
01:50 it's not a massive catastrophe.
01:52 But when you're talking about digital superintelligence,
01:56 I think, which does pose a risk to the public,
02:00 then there is a role for government to play
02:02 to safeguard the interests of the public.
02:04 (whooshing)