View source | |
# 2025-09-12 - Brewster Kahle On AI And Advertising | |
## Or How I Learned To Stop Worrying And Love The Bot | |
Recently i saw the EFF interview with Brewster Kahle about how to fix | |
the Internet. | |
Building and Preserving the Library of Everything | |
> AI is absolutely critical and actually has been used for, well, a | |
> long period of time. You just think of, how does the magic of | |
> Google search happen, where you can just type a few words and get | |
> 10 links and several of them are actually really quite relevant. | |
> ... | |
> And these materials are absolutely relevant now with climate change | |
> in our new environments because, well, things are moving. So the | |
> pests that used to be only in Mexico are now in Louisiana and | |
> Texas. It's completely relevant to go and learn from these... | |
Interesting to pose AI as a solution to climate change when it CAUSES | |
climate change. | |
Some time ago i watched the Nov 28, 1992 interview with Brewster | |
Kahle about WAIS, an early distributed search engine. | |
Interview with Brewster Kahle Regarding WAIS | |
I felt eerie watching this interview. It's as though AI were part of | |
the initial vision for creating Google Search and Internet Archive in | |
the first place. His answers seem prescient in many ways, talking | |
about technological convergence and people using their phone and TV | |
to post home videos. Sounds like Youtube and co. He was very | |
optimistic about how advertising would happen over the Internet. | |
Sadly, i think the human race has manifested something much darker | |
than his 1992 Utopian vision for the Internet. | |
What follows are relevant quotes from this 1992 interview. | |
36:20 | |
My personal goal out of this doesn't really have much to do with | |
electronic publishing at all. What I'd really like to make is a | |
machine that can answer the hard questions. A wise server. A server | |
that is somewhat the equivalent of the old style town priest or the | |
wise man on the hill. One you ask hard questions of. You might ask | |
"Should i go back to graduate school?" or "What's going to be | |
happening in pen top computers?" or "Is that a fluke? Is it a fad?" | |
That's [sic] the hard questions... I'd like to build that. | |
What does it take to make such a thing? Well, it requires a machine | |
that has a lot of experience. That has watched a lot of things that | |
have gone on. How people have tried things, and how they've worked, | |
and not worked. The priest, because of their special position in | |
society, gets to hear everybody's confessions, hear what it is | |
they've tried, what has worked and didn't work. And basically their | |
deal is that they won't go and blab it about. | |
I see electronic publishing as a major step towards making [this wise | |
computer]. A system for people to find what it is that they want and | |
pass this information around as a mechanism for getting this | |
[wise computer] going, in a way. | |
1:32:42 | |
One thing that i have a great deal of hope for is home publishing. | |
Instead of people just receiving information through their | |
television, through their mailbox, whatever; being able to put | |
information back out again through their computers, through their | |
telephones, through their television as a mechanism of encouraging | |
people in their homes to start producing things: little clips, home | |
videos, musical gigs, their ideas on whatever, can be packaged and | |
distributed from their homes. | |
1:35:53 | |
You asked "What is the role of advertising in the future of | |
electronic publishing?" Good question! One thing that I hope is | |
that advertising in its current form, if you can call it advertising, | |
because it's information that you don't want: it's sort of barraged | |
at you. It's usually done really inefficiently: you get all this | |
stuff in the mail... If there were advertising that you wanted, you'd | |
call it information. Like the Yellow Pages is *solid* advertising. | |
But [suppose] you care about auto parts [and] you'd really like to | |
know about what it is that product's offering you... | |
So there's a distinction between advertising and information: | |
advertising is those annoying things you get in the mail that aren't | |
targeted for you, but if those are targeted for you correctly, you'd | |
care. And if you were looking for it, then you'd actually want to | |
hear what the producer of the thing thought you should hear about it | |
and put it in the best light. | |
In the electronic world you have better filtering devices, it's sort | |
of like having a smart mailbox. Your mailbox... knows what | |
it is you are interested in and it throws out all the crap that | |
shouldn't be there but would actually go out and pursue the | |
information that you are interested in. So if you're about to go buy | |
a new VCR or a new camcorder or something, go out and probe around to | |
all those manufacturers, find out that information, correlate it with | |
whatever people have said, whether those were lies or not, how others | |
held up. Collect that information from a variety of sources, and put | |
it together so that you can view it. | |
I think the barrage style of advertising could really diminish in an | |
electronic environment based on the smarter terminals that we'll have | |
that will help correlate and filter out the crap. So is that good or | |
bad? I'd say it's good. | |
Most producers of widgets would really like to spend a lot of time | |
and money on finding those people that are thinking about their | |
product or thinking about going to a competitor, and are willing to | |
give free samples, give a trial run... [rather then spend a small | |
amount per person in a broadcast, they'd rather spend a lot more on | |
people who are more likely to follow through]. Electronics can help | |
producers find the right customers and that just makes the whole | |
thing more efficient. The stupidity of current advertising i believe | |
is an artifact of mass media. | |
p.s. | |
A relevant quote: | |
The right would have loved John McCarthy who coined | |
"artificial intelligence" largely so that he could avoid dealing with | |
Norbert Wiener. Not surprisingly he was also contemptuous of | |
Joseph Weizenbaum's critique of AI, writing: | |
> This moralistic and incoherent book uses computer science and | |
> technology as an illustration to support the view promoted by | |
> Lewis Mumford, Theodore Roszak, and Jacques Ellul, that science has | |
> led to an immoral view of man and the world. I am frightened by its | |
> arguments that certain research should not be done if it is based | |
> on or might result in an "obscene" picture of the world and man. | |
> Worse yet, the book's notion of "obscenity" is vague enough to | |
> admit arbitrary interpretations by activist bureaucrats. | |
From: https://post.lurk.org/@loriemerson/115198157075352690 | |
And from the comments: | |
> Weizenbaum's argument that computers should not be given tasks | |
> requiring empathy because computers are incapable of empathy seems | |
> fairly coherent to me... | |
This gets to the heart of the matter. If i were going to turn to an | |
external resource for wisdom that fundamentally lacked empathy, then | |
i may as well cling to a wire-monkey mother. | |
tags: political,technical | |
# Tags | |
political | |
technical |