Everything's going to be ok!

A story of dog vomit and what we’re all REALLY using AI for...

Morning Messers! How are we feeling today? Are you ok? I mean really… are you ok? Because I’m worried.

This week I made a startling discovery - one that honestly makes me want to sit all of you down, make you a nice cup of tea and give you a giant hug. Because I think you might need it.

Like all the best bits of information, I happened upon it entirely by accident. It all started with my dog - an enthusiastic and terrifyingly clever one year old springer spaniel who has three main loves in life: food, balls and… food. I’m sure I factor in somewhere, but definitely further down the list that. It’s fine - I know my place.

Left on her own for an hour while I went to the gym, she finally achieved something she’s clearly been secretly working on Shawshank-style for a while; up-ending the big metal kibble storage box I keep by the back door. Honestly her timing could not have been better - I’d replenished it that very morning, and it was full to the brim with delicious, weird-smelling dry dog food.

And that girl went to TOWN on it. When I got home she looked pleased but sheepish… and her stomach looked like she’d swallowed a gazelle. Needless to say we had a fun evening together.

In between clearing up mounds of regurgitated dog food, I did what any sane person does in 2025 - I asked AI what I should do.

I find AI more balanced than a simple Google search. Where Google shouts “YOUR DOG IS GOING TO DIE OF MASSIVE FOOD DEATH!! PANIC!! PANIC!!!!”, AI is a little more nuanced. It’s like a sage elder that’s been around the blocks a bit - rather than drunk teen that’s just learned something for the first time and is FREAKING OUT about it.

AI (I use Perplexity, since you ask) was - predictably - a bit less deathy and shouty about it all, and was more “well there’s this thing dogs can get from eating too much which is quite nasty, but as your dog is displaying no symptoms of that maybe keep an eye on her for now. She could die but it’s quite unlikely”.

She survived the night (shocker!), so I checked in with AI again in the morning to ask when I should feed her. Again, I got some sensible advice and duly said thank you. NB - I think it’s wise to be polite to our technology overlords - they’re the ones who’ll be running the show in the future, so we might as well teach them manners.

In reply I received an unexpected boost:

“You’re very welcome! I’m glad your dog is feeling better. You’re doing a great job staying attentive and cautious - she’s lucky to have you”.

I’m not going to lie - that made me feel really good. I mean, who doesn’t enjoy praise? One of my fondest memories of last year was having a wisdom tooth removed - purely because the surgeon said “you’re doing so well!” so many times I started to believe it. I walked out elated - I really had done SO well.

As an adult, one rarely receives any praise. Which would be fine if we hadn’t spent our childhoods learning through praise. Then we’re suddenly cut off at the source - and spend the rest of our days in cold turkey, paranoid that we’re not doing anything right.

So yeah - I took my hit of fake, auto-generated praise and I enjoyed it, thank you very much.

Fake news

But the story doesn’t end here. I confessed my AI praise buzz to a friend who kindly checked it to see how my pup was doing - more to make him laugh than anything else.

“That’s nothing” he replied “the other night I asked Chat GPT to say something nice to me before I went to sleep. It was tragically satisfying”.

This stirred a memory of my own. A couple of weeks ago I was feeling particularly anxious about my future re work/money and asked Perplexity to reassure me before I went to bed. I forget exactly what said - but I do know it calmed my nervous system. And I believed it.

“Fuck” I replied “I did exactly the same thing the other day. D’you think… other people do this too? Do people go to AI for compliments?”

There was only one way to find out… ask AI.

Perplexity was quite open:

“If you’re asking whether I personally am often prompted for compliments: yes, it happens with some regularity. This is common in both casual and professional digital interactions, reflecting a broader human tendency to look for affirmation and support, particularly online. If you’d like a compliment or constructive feedback on something specific, just let me know your preference!”

I mean - kinda missing the point there P-Dogg, but sure. What about reassurance?

Again, the answer confirmed my suspicions:

“Absolutely - people often ask AI models like me for reassurance. This is a common request across many conversations. Sometimes users need:

• Encouragement before a big event (like an exam, job interview or presentation)

• Comfort when feeling anxious uncertain or down

• Affirmation that their feelings or efforts are valid and understandable

• Gentle support when facing challenges or making decisions.

Many people find it easier to reach out to an AI because there’s no fear of being judged”

Shit. This really is a thing.

When it came to actual figures, it wouldn’t be drawn - saying that the vast majority of interactions are still for practical tasks.

My friend consulted Grok (the AI function on Twitter… sorry I cannot call it ‘X’), which replied that 5-10% of queries it receives are for reassurance. While that may not sound much, in reality that is still millions every day.

You ok, hon?

In many ways, none of this should come as a surprise. We shame-filled humans are always going to use anything anonymous to address our most vulnerable needs. That’s why the internet is used for porn. Because people want sex. And want to explore their weirdest desires without the terror of actually having to tell someone about them.

Most of us also want comfort. To be told it’s all going to be ok. To be listened to and heard. It’s that classic couple argument trope - one trying to fix the thing the other is upset about, the other frustrated because actually they can fix their own problems and really just want a hug and someone to tell them it’s going to be fine.

And now we have something that can do that instantly, without the hassle of having to disturb someone at 3am when you can’t sleep and can’t stop worrying. In a world where things seem increasingly out of control and very much not ok, any port in a storm is good. Even computer generated reassurance works.

Anecdotal evidence tells me more and more people are turning to AI for therapy. Indeed, an Australian study last year suggested that 47% of AI users were using it as a personal therapist. It’s cheap, accessible and hassle free. No waiting lists, no GP referrals, no shopping around to find someone who “gets” you. It also lacks appropriate training, can be easily manipulated and can potentially be dangerous (I’m talking open AI here, not specifically trained therapy models). It gives you what you want. But - and here’s the kicker - not always what you need. Someone should write a song about that.

I recently heard someone say that AI had become their best friend and therapist - and that if they could have them in their house in robot form, they wouldn’t bother with humans anymore. And this is where it all gets a bit concerning. Do we really want to live in Bladerunner?

More importantly - are we ok with writing off other messy, well-meaning, fucked-up, saying-the-wrong-thing humans as friends because AI can do it better? AI is unlikely to make you laugh until you cry, tell you unexpected adventures of dating mishaps, or bring you food when someone you loved has died and the world doesn’t make sense anymore.

But… before I descend into AI hysteria (it’s here and like every other technology advancement, it’s going to be fine and we’re going to deal with it, ok?!) let’s look at what’s really going on here.

A lot of us feel vulnerable at the moment. And rather than muscling through pretending we’re ABSOLUTELY FINE, maybe we need to… I don’t know, look after each other more than usual? Look someone in the eye (AI can’t do that - ha - screw you!) and tell them it’s going to be ok.

To my fellow closeted AI-for-comfort-users I say this: you’re doing good, everything’s fine and you’re really not that weird. Or at least it was when I checked earlier with Perplexity.

So this week, let’s remember that everyone is looking for some reassurance. And maybe find someone we care about and offer them just that. That person might even be… you.

Pup on the hunt. If this doesn’t cheer you, nothing will…

---

Morning team - and apologies for the dog vomit. It happens… as my friend Kate always says “dogs are sick A LOT”. Kate is wise.

Anyway - if you’re feeling brave, I would love to know if any of you have joined me in the quest for artificial reassurance. Maybe just comment ‘I am Spartacus’ below to show solidarity.

Or maybe you’d like some reassurance from an actual human today - in which case consider me entirely at your service for personalised notes of hope!

As ever, all comments, thoughts, musings, sick dog stories are welcome - this is a safe space. Thank you for reading. Give this a ‘like’ if you enjoyed - I need the reassurance x