Calling John Kennedy's (US Senator La.) Bluff

In the recent Senate hearing on regulating artificial intelligence (AI) Senator Kennedy, who always presents himself as an "awe shucks" sort of fellow, said this:

"This is your chance, folks, to tell us how to get this right. Please use it. ... Talk in plain English and tell us what rules to implement."

It isn't clear to me who the "folks" he is commanding to tell him "what rules to implement" are, but OK, Johnny, I'll tell you something.

I'll tell you how to find out a lot of what you need to know to figure it out for yourself - you were elected to do something, after all (one would suppose) so I'll tell you where to look.

First, though, I need to observe that I would expect that a Vanderbilt and Oxford educated fellow would want to craft the laws himself, not be told "what rules to implement".

I, personally, have no idea "what rules to implement".

I didn't go to Oxford.

But I still think a lot, despite advanced age, and I have a pretty good idea of what those who are charged with "what rules to implement" - such as you - ought to be worrying about, concerning AI.

So I can't tell you where to go to be told "what rules to implement" but I can tell you where to go to find out all that you need to know to at least begin to figure out for yourself "what rules to implement".

I hope you can find time to do what I am telling you to do.

But then, being the "awe shucks" sort of guy that you are takes a lot of bandwidth.

So, we shall see.

Senator Kenedy: do the following:

Watch a recent 8 minute interview on PBS Newshour of Geoffry Hinton.

Watch a recent EmTech Digital 20 or so minute interview of Geoffry Hinton.

Watch a recent CBS Saturday Morning 70 or so minute interview of Geoffry Hinton.

Watch AI and the Future of Humanity, Yuval Noah Harari at the Frontiers Forum - 20 or so minutes.

Read The Alignment Problem by Brian Christian.

Read Artificial Intelligence A Guide for Thinking Humans by Melanie Mitchell.

Read the three articles about AI in the 22 April edition of The Economist.

Do some chats with the new Chat GPT enabled Bing.

Pull your head out of wherever it seems to be.

I have done all of those things, including the head thing.

But I'm just a drone out here who, having pulled his head out, periodically wakes up and pays attention to stuff until I go back to sleep; I'm not a brilliant thinker and classic wordsmith like you, and this watching and reading list is a pittance of what is available for deciding "what rules to implement".

However, I have to assume you haven't done any of the things I am suggesting to you, or you wouldn't want to be told "what rules to implement".

So, get crackin'.

POST SCRIPT

I decided to email the link to this post to Senator Kennedy.

It turns out that that's not really possible.

I went to his web site.

No matter how many times you click on the "email Senator Kennedy" link, you end up with no email link.

If you keep trying other ideas and succeed in getting close, you will be trapped in some kind of online form (I had to pretend that I had a legislative proposal to get that far).

There is a "Topics" drop down menu.

Here are the allowed topics starting with "A".

Since Artificial Intelligence isn't on his "topics" list I guess we must assume that he isn't very interested in that "topic", his desire to be told "what rules to implement" notwithstanding.

I guess we better hope somebody smart tells him "what rules to implement".

Maybe GPT4 or GPT5?