WWN24 : Dorkbot 5000

One of the best things we done bought our kids was a Yoto player.

For those as don’t know, those are a simple audio player.

They play audiobooks and music.

But the genius of the Yoto player is that the books and music come on cards, drop the card in the top and it plays. Totally simple, totally safe, no screens involved.

My oldest daughter loves it and regularly listens to books like Stuart Little or the Secret Garden on repeat.

(Plus so much Enid Blyton that they all come out with classic 1940s phrases like “Oh, Mother and Father will not be pleased…” and so on. Hilarious from a three year old.)

It works wonders for her vocabulary, comprehension, focus etc. and keeps her away from toxic screens while still letting us breathe.

Well, we had to get my second daughter one as well.

Only, we got it and a few weeks later the buttons broke.

Not to worry, I finally got around to reporting it to customer service and got this reply…

Now…

…I dunno about you.

But to me, if the buttons on your device are completely broken and don’t work, the correct customer response is “send it back and we’ll replace it” and then they can get the details from there.

Then again that’s not the point of this newsletter. That’s coming in a moment.

(Eagle-eyed readers will have already spotted it.)

I sent the information and got this response:

The eagle-eyed among you will perhaps have noted what’s triggered my frustration…

There in the sign-off.

“This response was created by AI.”

I mean, I already knew that.

Either it was AI or someone who speaks English as a foreign language and learned primarily from reading HR manuals.

It’s given away by phrases like “It’s important for all features to be functional, especially when the product has been well cared for” which no human being would ever knowingly type.

(I mean, duh. Right? That’s not something one person says to another person to comfort them, unless said first “person” is a robot.)

Neither is “review the malfunction and expedite a solution”, which is robot-speak for “figure out what’s wrong and fix it.”

But it’s all part of this brave new world the sexbot-humpers at OpenAI and online are trying to promote. A world where instead of sending an email that gets opened by a human being the next day, who uses their actual intelligence and fixes it…

…you have to play “jump the hoops” with a robot instead.

(Everywhere does this now, it’s impossible to get customer service without wasting an hour of your life with a chatbot first. I hate it.)

None of this, however, is really about Yoto and their customer service choices.

I’m sure it’ll be resolved swiftly and happily.

But look back at that AI “writing”…

And now let’s consider how a human would write that:

See how much better that is?

(And shorter? And more human?)

Now, I’m sure that the AI shills will rage that it’s all about the prompts and just you wait two years (another two years, that is) and so on.

Maybe.

I’m sure this isn’t a top-end AI system.

But then again…

…every single AI piece of writing that I’ve ever done seen is like this. It’s all so…

…Soulless and robotic.

(Weird, how soulless robots sound like soulless robots, huh?)

And it’s a widespread problem.

A couple weeks ago I hopped on a call with my friend Dino (a.k.a. ArtofPurpose) to school his Masterclass 24/7 members on tweet writing.

(That’s an affiliate link FYI because I endorse it wholeheartedly for those looking to build an audience.)

All part of starting to hype people up for the next round of Create, Publish, Profit, which kicks off tomorrow.

(Affiliate link for that coming at the end.)

So we hopped on a video call, I lit up a pipe (Hoggarth's Coffee and Caramel), poured a whisky (Blair Atholl 12), and we went at it.

It was a blast, we got great feedback.

"Awesome feedback, thanks guys"

"this is great"

"That’s a killer hook, I love it"

"Thanks."

"Thanks this was so helpful."

"Thanks for the feedback. Harsh, but accurate"

Etc.

And I’ll be doing more of that in CPP for those that are signed up.

But there was one poor guy who we roasted. I’m not going to identify him here because he's a solid account doing good work and I like the guy. Real nice guy.

(Nor am I going to post the offending thread because it's on the timeline and you could find it.)

But Dino put it in the content creation room in Masterclass, and we started to read it.

A minute later I have a suspicion.

And then a minute after that, Dino asks:

"What is it with these blocks of text? It reads like something that… Did you use ChatGPT on this?"

"I use chatgpt as a baseline, yes."

Dear reader, let me tell you…

Sparks flew!

Because Dino is equally mad about people using CrapGPT to 'write' for them as I am.

And he’s right, because just look at those robo-service messages I got and tell me that looks good for your brand as a creator of any kind…

Truth is, you cannot use CrapGPT or any kinda AI tool to do your writing for you!

People will be like "Oh, but I keep saying people say that it's gotten good."

Yeah, people who can't write think it's gotten good.

Because what it's gotten good at is churning out bland, forgettable, cookie-cutter jargon-filled content. Content like that thread or customer service replies.

"Look! It put words on the page! So cool!"

To which all I can respond is:

Shut up dork!

Those are dead words. Dull words. Downright daft words in some cases.

And yet people insist on pretending the Robo-Emperor has clothes on.

Even creators who I respected pull this crap recently. Normally, because they want to make a buck selling you a dumb AI workflow/system/tool that can write your emails for you in five minutes.

Clear off.

I've never once read one of these oh so good sample emails that is meant to be a killer example and not known it was AI within about two sentences.

It all sounds so bland. So forgettable.

So like the customer service messages I got this week.

A while back, the internet market Justin Goff ran a promo on that basis with three passages and asked people to guess which one AI had written.

The answer was all of them, gotcha!

Except I'd called that…

…within ten seconds of looking at it.

They all had that same dead, lifeless robo-voice that only comes from getting AI to 'write' for you.

In fact, our pal within Masterclass who was getting roasted confessed that he'd taken a two page transcript he wrote himself myself and had CrapGPT "rewrite" it and that's how he got the thread we roasted.

He posted a version of that original later, and believe me. It's a thousand times better. Because it sounds like a freaking human wrote it instead of Dorkbot 5000.

I'll have a lot more to say on developing a voice that injects flair and personality and can never be duplicated by robots soon but for now let me finish with this:

Dear reader, if you want to sound like a robot, get a robot to write for you.

Otherwise?

Do it yourself.

And if all this offends you because you have some mistaken notion that you can prompt your way to a personality?

Then the unsubscribe button is waiting to be used.

It's down there below the fold, right after I finish this email.

(If you're struggling to find it, ask CrapGPT. I'm sure it can tell you where it is.)

But if you're still here, you wonderful luddite, then may your pipe and your prose be personal,

Yours,

James Carran, Craftsman Writer.

And for anyone who is definitely totally a human and not a robot, maybe you'll consider our expensive but AI free cohort on building an audience and a personal brand (with an emphasis on the personal).

It's the last round this year, and maybe the only one I'm running my particular session on.

And it’s only for those who are serious about committing to building a business based on X. I attended the very first prototype cohort that became CPP and I’ve helped with every cohort since then. I’ve seen businesses blow up as a result, I’ve also seen lots of people have a great time and then disappear because they never really committed.

So no hard sell from me, as always.

But if building a business on X interests you, check out CPP through my affiliate link here.

You’ll help fund the Write Way at no extra cost, and I’ll also throw in a substantial bonus.

I have two courses that you currently cannot buy anywhere else that you’ll get free.

Voice Box (about to launch in beta), which builds on the material I’m covering in my seminar in CPP about being Utterly Unique. And Effective Editing, which is my flagship course (currently in beta) about editing effectively, building on the tactics and tricks I’ll be using in CPP to edit your content to make it perform better on the timeline.

And I’ll also add another bonus that I call “double your content”. Send me two weeks worth of tweets (up to 50) and I’ll go through and rework them all with new ideas and formats, write new tweets etc. so that you have twice as many to test on the timeline.

Which is a lot of bonuses and if any of that interests you, head up to the link and decide before the cohort starts.

You have less than 24 hours.

James Carran, Craftsman Writer

fin

And fin also for the fine folks in Craftsman’s Corner who keep the lights on around here.

No, I’m not abandoning you. But this week I’m going to pop Craftsman’s Corner into The Weekender instead, so look for that on Saturday. And if you’re not already inside, go here and upgrade:

getpaidwrite.com/upgrade

Of course, The Weekender is never published to the getpaidwrite site, so if you’re not upgraded before I send it on Saturday, no Craftsman’s Corner for you.

(and with that, it’s really fin for you all)

P.s. Carran’s Cabin subscribers, pay attention to how I reworked a couple of emails to you into the meat of this email. Lesson in there. Ghostletter Seminarians will also know what part I changed and why…