Script from my national radio report yesterday regarding Google AI Overviews
This is the script from my national radio report yesterday regarding
Google AI Overviews. As always, there may have been a few very slight
word variations as I presented this live.
- - -
So yeah, I've previously mentioned the controversies spinning up when
Google began recently pushing their Search AI Overviews out to all
Google Search users instead of just users who had opted in from Labs,
and since then the situation has really gone rather ballistically
downhill. Google's latest move in the AI hype race is to push out
AI-created "overview" answers at the top of many search results,
causing a bunch of problems.
Those AI overview answers can take a lot of space, so users must often
scroll down the page to find most of the links to sites where Google
got the information from in the first place. Some of these AI Overview
answers do theoretically show links with them, but typically they're
in a tiny font, blurry unless you have great vision, often in pastel
colored boxes making them even harder to read. Google claims there
are still lots of users clicking through, but many sites have
apparently had their traffic drop massively since Google started AI
Overviews.
Another problem is that AI Overview answers are often just plain
wrong, or incoherent, and in some reported cases potentially
dangerous. So as all this has been noticed the amount of basically
public ridicule aimed at Google recently has been dramatic, with some
observers saying that Google has basically broken search and by
starving sites of traffic could effectively break much of the Web,
especially for smaller sites. Google is already that target of
government antitrust actions and now these AI Overviews and their
effects are triggering new calls for government regulation of Google's
activities, given their enormously dominant position when it comes to
search activities.
And Google's reactions to the criticisms have been, well, interesting.
They insist most of the AI Overview answers are good ones and claim
that these are the answers we should concentrate on and not the
inaccurate or wacko answers. And they also seem to be doing a bit of
"victim shaming", claiming that people just shouldn't be asking
questions of Google that will cause their AI Overviews to give bad
answers. Which seems like rather a public relations kind of viewpoint
that pretty much completely misses the point of why this situation is
such a problem.
Nonetheless, some days ago I noticed that the number of queries to
Google search that were getting AI Overview answers seemed to have
fallen significantly, and in more cases there were simply only the
normal blue links that Google so much seems to want to de-emphasize
now. Or blue links with the old-style "featured snippet" at the top
with a direct link to the site from which the snippet was sourced. And
Google then apparently as part of their attempt to justify this AI
Overview mess said that (for now at least) they've cut back the number
of AI Overviews, so that they'll mainly appear on questions where they
are the most helpful in Google's view, rather than seemingly on so
many queries as it was originally. So my sense that AI Overviews were
reduced appears to have been accurate.
Google does now seem to be reacting to public concerns about AI
Overviews, but whether this is just a pause that will be reversed in
the near future or not is a different question entirely. I have no
expectations that Google is really backing off in the long run on
various of their AI efforts, even the particularly misguided ones.
So it seems most likely that we're going to be back discussing these
and other Google AI-related issues frequently going forward, absent
some drastic change in Google management's philosophy, and I
definitely wouldn't suggest betting the farm that such a change is
going to happen anytime soon, or quite possibly, ever.
- - -
L
- - -
--Lauren--
Lauren Weinstein
[email protected] (https://www.vortex.com/lauren)
Lauren's Blog: https://lauren.vortex.com
Mastodon: https://mastodon.laurenweinstein.org/@lauren
Founder: Network Neutrality Squad: https://www.nnsquad.org
PRIVACY Forum: https://www.vortex.com/privacy-info
Co-Founder: People For Internet Responsibility
Tel: +1 (818) 225-2800
_______________________________________________
google-issues mailing list
https://lists.vortex.com/mailman/listinfo/google-issues
