Odie Lives! Poking Search Engines With A Stick

I was just looking for a few original images of Odie, from the Garfield comic strip. DDG gave me a load of crap.

So I typed this:

odie garfield original NOT PLUSHIE JUST ORIGINAL COMIC STRIP YOU STUPID AI SHITHEAD

And it gave me what I wanted.

Rule 1: DO NOT BE POLITE.

You go Douglas! I'll try that next time stupid stuff is in my search results.

I tried Comic Strip odie and got lots of comic strip featuring odie with Mr Google.

Jon Arbuckle's pet dog (formerly Lyman's). He lives in Jon's house with Garfield, who relishes in kicking him off of every conceivable surface.

(While reading this, remember that Odie is a dog, and thus feats that would not be impressive for a human are impressive for him. Only feats from the 1978 newspaper comic strip will be in this RT)

https://www.reddit.com/r/respectthreads/comments/1e7ywzs/respect_odie_garfield_newspaper_comic_strip/

Good to know.

It does make me extremely frustrated when a search engine gets in the way of meaningful searches. DDG (and its parent, Bing I think) automatically assume that if I'm searching for something, I want to BUY something instead of doing research. At least that's my impression (hence the temper tantrum above).

Google results are better but the creep factor is ... way too creepy. I recall there are "Google anonymizers" (?) listed in another thread (started by me, again ranting). I really need to look into those.

I wish there was a way to turn off the AI answer at the top of google searches. I hate how my brain is like, "Oooh, answer!" for the first second I see it, then I have to scroll down and check actual websites. Just give me the actual websites first! If I wanted AI answers, I'd ask for them. I asked for search results, not "answers"!

I was doing a google search earlier today, and I realized part of the reason that I end up reading the AI answer, even though I don't want to: It takes a few seconds for the rest of the page to load and allow me to scroll.

I am stuck looking at the AI result until I can scroll. So, I read it while I wait.

I don't know if this happens simply because of my slow internet, or if it happens to everyone and is intentional on Google's part to get us to read the AI answer.

Things like this have me convinced that AI is just too stupid to take over the world.  Artificial Idiocy.

Last night I looked up some information. The AI answer popped up. It didn’t look right. I clocked onto the expanded answer …it was completely different and contradicted the “ headline answer”

Nicole Alderman wrote:I don't know if this happens simply because of my slow internet, or if it happens to everyone and is intentional on Google's part to get us to read the AI answer.

It's not your Internet speed. It is Google strategy.

The black humour part of this adventure is I was hoping to cobble together a meme for our work team, adding "Odie the AI Intern" (endlessly eager, not housebroken, watch where you step). *facepalm*


Bad search engine Odie! Bad dog!

Ellen Morrow wrote:Things like this have me convinced that AI is just too stupid to take over the world.  Artificial Idiocy.

AI may be dumb as a rock, and it may be measureably dangerous.

https://arstechnica.com/ai/2026/04/anthropic-limits-access-to-mythos-its-new-cybersecurity-ai-model/

I hope this rather heavy quote meets the standard for "fair use" (such a quaint notion).

Mythos has been in use with partners for several weeks. Although it is a “general purpose” [cybersecurity AI] model with wider capabilities, it is the first time the company has limited release of a model due to its capabilities in cyber security.

Anthropic said the software can identify cyber vulnerabilities at a scale beyond human capacity, but it could also develop ways to exploit these vulnerabilities, which bad actors could use.
...
In recent weeks, Mythos has identified thousands of so-called zero-day—previously undiscovered—vulnerabilities and other security flaws, many of which are critical and have persisted for a decade or more.
...
At one point, Anthropic found that it had escaped its so-called sandbox environment—designed to prevent it from accessing the Internet—and posted details of its workaround online.

Anthropic acknowledged it demonstrated “a potentially dangerous capability for circumventing [the company’s] safeguards.”

Folks, you just keep on building your skills and gardens, and hang onto your off-grid capabilities as your ace-in-the-hole.