Google

On our encroaching “AI” dystopia

I put “AI” in quotes because it’s not really about intelligence at all, people just glommed to the term because it:

  • Already existed
  • Sounds futuristic
  • Sounds high-tech

Google released an ad that has been airing during the Paris 2024 Summer Olympics (as an aside, this is the first Olympics that has been known probably as much for its meme generation as the athletic competition) which showed a father helping his young daughter write a fan letter to Olympic athlete Sydney McLaughlin-Levrone–by using Google’s Gemini AI.

The ad (and Google’s defense of “it tested well”) is wildly tone-deaf and a prime example of how people don’t want AI to work. Leave the creative stuff to humans, AI can handle the drudgery. And yet a lot of the big AI push is the exact opposite.

Ars Technica story: Google pulls its terrible pro-AI “Dear Sydney” ad after backlash

In the comments to the above story is the following comment, which resonated with me:

A growing number of people expect the AI bubble to burst, it’s just a question of how soon.

I’m thinking it will be sooner rather than later. We’ll see what happens by the end of the year!

Google’s attempt to enrich Google through “AI”

I feel like “AI” always needs to be in quotation marks, because while it is clearly artificial, there is no apparent intelligence involved. It’s all just an elaborate guessing game based on a giant pool of answers that could be accurate, inaccurate, made-up, sarcastic, or some enticing combination.

Google is now rolling out an “AI Overview” in its search results, a furthering of its efforts to keep everyone on its search page (to show them ads and make $$$), with the side effect of killing off the rest of the web as all other sites slowly starve for traffic, revenue and everything else.

It is easy to find examples of this overview being hilariously and sometimes dangerously wrong. Google appears to be fixing the most egregious examples, no doubt by coming across them in their own searches and then fixing the results manually (ie. with an actual human). If this is their plan, they are going to need a lot of humans.

Here’s just one story from Ars Technica and one screenshot (below) illustrating the whole big pile of nonsense. I really don’t think this is going to improve over time in any significant way. Since this is Google, I’ll go further and say AI Overview could eventually end up in their graveyard along with the hundreds of other things they’ve killed.

GIGO

This may or may not conclude AI Week on the blog.

Googled, The Verge Edition

No, I have not Googled

Google may have a monopoly on search engines, but that doesn’t mean writers at major tech websites should be greasing the way for Google by using “Googled” as a synonym for “searched”. This is the kind of thing that used to give Xerox fits when people said they “Xeroxed” something instead of “photocopying” because if Xerox didn’t fight back against the use of its name as a verb (or noun), it could eventually be declared generic and fall into the public domain.

On the other hand, Google probably doesn’t care because they are a mega-company and don’t fear consequences over things like trademarks or copyright. They know they’re in no danger of “Google” becoming a generic term, and actually like it when people say “Googled” because it further cements “search” as being synonymous with Google.

And this is why writers should do better than to help Google along. Other search engines exist (I use DuckDuckGo and Kaigi) and it’s presumptuous to assume everyone uses Google.

(BTW, the editorial is otherwise well-stated and worth checking out!)