Have you heard of the Great AI Open-Closed Soap Opera, starring Sam Altman, Elon Musk, Mark Zuckerberg, and a peanut gallery that just keeps getting louder?
This is truly the blockbuster sequel nobody had asked Netflix to produce, but everyone secretly wants to watch, because heck, what else you gonna do besides “work” (and in your case, sit on the couch, surfing da weeb).
This, my friends, is “AI Openness 2: The Electric Boogaloo”.
And yes, it’s just as corny as it sounds.
Sam (the Altman, yes), recently took to Reddit for a “no-holds-barred” Ask Me Anything session. Sounds bold. Except the big question everyone wanted an answer to was: why did OpenAI, with “open” in the actual name, decide to put its AI models behind lock and key?
In the AMA, Altman basically said, “Look, you guys, we’re just too good at this AI thing. And, you know, safety”. Which sounds like, “We’re keeping the goods because we don’t trust you with them”.
Wud? Safety first, right?
This is being invited to a potluck and then being told you can only have the salad because the host is worried you’ll choke on the main course.
This whole piece of comedy is shaping up to be a full-on episode of The Muppet Show, with Statler and Waldorf sitting up in the balcony as our resident judges, ready to tear it apart with their classic one-liners and relentless sarcasm.
Before we start!
If you like this topic and you want to support me:
- Comment on the article; Google appreciates that and it will really help spread the word 📢
- Connect with me on Linkedin 🔗
- Subscribe to TechTonic Shifts to get your daily dose of tech 💉
- For my Dutch friends – register for the webinar: The Evolution of AI 📈
Mr. Statler, Mr. Waldorf, thank you for joining in.
Statler: “Joining in? Oh boy, they must be desperate if they’re asking us to join anything!”
Waldorf: “Yeah, desperate for quality entertainment! We’re here to add a little class… by dragging it down!”
Statler: “What are we reviewing this time? Another company trying to define ‘open’ without actually opening anything?”
Waldorf: “Sounds like a classic! More closed doors than a Muppet backstage rehearsal!”
Statler: “So, we’re the experts here, huh? Guess they ran out of people with actual insight!”
Waldorf: “Let’s get this trainwreck started! Who needs logic when you’ve got sarcasm, right Statler?”
Statler: “Right! It’s the only thing truly open around here!”
Meanwhile….
Elon Musk, who co-founded OpenAI back in the day before, let’s say, creative differences made him bolt for the door, has been tweeting up a storm about OpenAI’s tragic transformation.
He is like your ex who keeps bringing up how “they used to be all about love and trust” but are now just “a bunch of capitalists with a marketing gimmick”. In Musk’s words, OpenAI has morphed from “open-source & nonprofit” to what he calls a “capped profit” structure.
Translation for y’all n@@bs: “OpenAI is basically trying to wear a halo while also cashing in at the register”.
So Musk, being Musk, slapped OpenAI with a lawsuit because that’s just how he shows he cares. He claimed they abandoned the whole “open” concept and sold out. And OpenAI, being quite open about Musks correspondence, responded with actual emails from when Musk was there, showing him agreeing to make things less open.
Oh, the irony.
The best part is that Musk’s entire response was a one-word “Yup”.
I think it’s just his crooked way of saying, “Sure, I’m part of the problem, but it’s more fun being the solution now”.
Statler: “So let me get this straight. Musk complains they’re not open, but he also said it was fine to not be open back in the day?”
Waldorf: “Exactly. It’s a classic ‘Do as I say, not as I do’ situation. Or maybe ‘Do as I vaguely imply in a cryptic one-word email, not as I tweet.’”
Enter the Zuck
Just when you think this OpenAI drama can’t get any juicier, along comes Mark Zuckerberg, a.k.a. the Zuck. And he brings along Meta’s latest AI models, the Llama 3.2.
And the Zuck swears by open-source, as you know.
He is even calling it the path to a “safer” AI future.
Of course, there’s a big twist, because there always is: if you have more than 700 million users, Meta’s not just going to hand you the Llama. naaahhhh you’re gonna need a hall pass.
Because who doesn’t love a little selective openness?
For those of you scratching your heads wondering what on earth that actually means, let me explain: Meta’s Llama isn’t really “open-source” in the way you would expect. Sure, they call it open-source, but it’s more like a conditional open-source.
If you’re a small player, come on in.
But if your user base is a bit too big, you’ll need to file a formal request, pass Meta’s vibe check, and maybe send in a resume or two. And then, just maybe then, you can become a member of their secret club where you’re only welcome if you’re not too popular.
A.k.a. “open with a twist”!
Meta’s response was almost poetic in its corporate nonchalance.
“There’s no single open-source AI definition”, they said, and they were basically arguing that the definition of “open” is open to interpretation.
The OSI (which stands for the Open Source Initiative), they rolled their eyes harder than a junkie who just got his fix, but Zuck’s team stayed unbothered. Their message was essentially, “Hey, let’s not get bogged down by technical definitions, peeps”.
It’s the classic “free admission…for everyone we like” kinda move.
Statler: “So Meta’s Llama is open, unless you’re a big player. That’s like inviting the whole neighborhood to a BBQ but slamming the gate shut if they brought too many friends”.
Walforf: “Yep. It’s ‘open’ like a VIP section at a stripclub. Come in, but only if we’re cool with it. Open-source exclusivity, the Meta peepshow style.”
Sam has a Second Brain
Meanwhile, Sam Altman is over at OpenAI trying to ease the backlash with a sprinkle of “don’t worry, be happy, and we’ll open source more stuff in the future, we promise”. This is a classic example of “the check is in the mail” reassurance.
What exactly does “in the future” mean, Sam?
Are we talking 2025? 2030? Or the year we’re all hooked up to Neuralink and every thought we have is instantly auto-Tweeted to Grok on steroids (read: XXX)?
To really make this a full house, OpenAI also casually mentioned that GPT-5 won’t be released this year. Shait… I just wrote in the AI | ML weekly, it will probably launch in December.
But, Altman also whispered that he’s actually using ChatGPT to answer some AMA questions.
You heard that from me first, my friends !
The CEO of OpenAI outsourced his AMA to his own Second Brain AI.
I like hired my kids once to do the garden because hey “they’re just better at it” (and cheaper).
I mean, if the AI is already answering questions about itself, we might as well let it run the company next.
Statler: “Wait, so Sam can’t be bothered to answer questions about his own company? He’s making ChatGPT do it?”
Waldorf: “He is also probably asking the chef how they made the soup, and they just hand you the recipe on a Post-it note. Doesn’t feel very ‘personal touch’, now does it?”
As if this rollercoaster needed any more loops, we’ve got three big personalities giving their opinions on the situation.
We’ve got Musk who is out there saying, “AI is supposed to be for everyone, like a free buffet. Why are we charging for the main course now”?
Altman’s response was “Well, someone has to keep the buffet in line, Musky”.
And the Zuck is all in with, “How about an open buffet, but only if you don’t bring too many friends”?
At the end of the day, the Altman’s promise of “openness someday” feels a bit like those “new Coke” promises. Remember new Coke? Nah, you were probably not even born, and I am talking Muppets, man…
Anyways, new Coke sounds nice, but will it ever actually come? As I am sitting here with my popcorn, watching der Altman, ze Musk, and the Zuck verbally joust, we’re just left wondering who gets to define “open” in this strange new AI world.
Well, maybe someday, when AI models are doing our taxes, managing our calendars, and yes, answering AMAs for us, we’ll look back on this debate and laugh…or maybe we’ll be too busy asking our AI assistants what “openness” means.
Statler: “Well, that was… something. We came, we saw, we criticized every inch of it.”
Waldorf: “Yeah, we gave it the full treatment. If they were hoping for applause, they’re in the wrong balcony!”
Statler: “So, OpenAI isn’t so open, and Meta’s open-source comes with a membership fee. You know what this reminds me of?”
Waldorf: “A locked door with a sign that says ‘Welcome’”!
Statler: “Exactly! It’s like calling yourself a vegetarian restaurant and only serving steak!”
Waldorf: “Or naming yourself ‘The Muppets’ and putting us in charge of quality control!”
Statler: “In the end, we learned one thing: these tech giants sure know how to dress up a closed door and call it ‘progress.’”
Waldorf: “Right, and we’ll be here next time to tell them they’re still doing it wrong!”
Statler: “At least until they close us off too!”
Both: “Ha ha ha ha!”
Signing off – Marco
Well, that’s a wrap for today. Tomorrow, I’ll have a fresh episode of TechTonic Shifts for you. If you enjoy my writing and want to support my work, feel free to buy me a coffee ♨️
Think a friend would enjoy this too? Share the newsletter and let them join the conversation. Google appreciates your likes by making my articles available to more readers.
Leave a Reply