Inside Meta’s attempts to play catch-up with AI

It is amazing to me how Facebook is still in Too Big To Fail land...in spite of their loser CEO not having a successful launch of a product since...well...we became Dictator in Chief of that company. He has almost as much of a Tony Soprano Touch as Trump does...except people keep handing Zuck mountains of money to just ineffectually set on fire, in spite of that.
 
Upvote
87 (87 / 0)
It is amazing to me how Facebook is still in Too Big To Fail land...in spite of their loser CEO not having a successful launch of a product since...well...we became Dictator in Chief of that company. He has almost as much of a Tony Soprano Touch as Trump does...except people keep handing Zuck mountains of money to just ineffectually set on fire, in spite of that.
It's pretty common for one hit wonder billionares to fail time and time again.
It's because they were lucky and in the right place at the right time.
They're not smarter than you or me. They have no great insight other than what they can glean from their business, That's why when they try other adventures, they fail.
 
Upvote
79 (79 / 0)

MrDweezil

Wise, Aged Ars Veteran
189
Aren't the models themselves are going to hit some level of "good enough" and become fairly commoditized, with the value coming from the tools built around them? The OpenAI/Google/Anthropic trio are going to get to that level before Meta does, so it seems like a pretty questionable use of investment.
 
Upvote
28 (28 / 0)

gtrawo

Smack-Fu Master, in training
51
Some employees have also protested company plans to install tracking software that would capture their computer usage in order to train AI models. Meta on Tuesday told staff in a memo, seen by the FT, that it would roll back parts of the plan following the backlash.
Maybe that’s where Microsoft’s Recall can find a home! In the sweat shops of the West.

Smh
 
Upvote
10 (10 / 0)

Kevin G

Ars Scholae Palatinae
1,488
Meta honestly knows that they don't necessarily have to close the gap because they have one of the largest data sets to build AI models off of. All that Facebook and Instagram data users have generated over a decade is still sitting in their data centers being scrapped by their own AI. The data set itself is a strategic advantage. Meta will continue to run their own AI using their own models based upon their own data. They are large enough that if they see a small upstart they can acquire them and then integrate the new IP with Meta's vast data repo.
 
Upvote
6 (6 / 0)
It's pretty common for one hit wonder billionares to fail time and time again.
It's because they were lucky and in the right place at the right time.
They're not smarter than you or me. They have no great insight other than what they can glean from their business, That's why when they try other adventures, they fail.
In fact, they're usually dumber. If you look at most billionaires the only thing separating them from you and me is that they had access to wealth. Because of that wealth safety net, they can try and fail repeatedly until they hit on something that works. The average person starting a business has only 1 shot, while billionaires usually start wealthy so can try multiple times to hit it big.
 
Upvote
33 (34 / -1)

NoSkill

Ars Praetorian
531
Subscriptor
They're not smarter than you or me. They have no great insight other than what they can glean from their business, That's why when they try other adventures, they fail.
The Wealth Cushion gives them access to innumerable resources to navigate failure. Bankers, attorneys and accountants to name a few on the inside, and consultants for specialized expertise and external events.
 
Last edited:
Upvote
28 (28 / 0)
It's pretty common for one hit wonder billionares to fail time and time again.
It's because they were lucky and in the right place at the right time.
They're not smarter than you or me. They have no great insight other than what they can glean from their business, That's why when they try other adventures, they fail.
It's actually worse than that. Yes, most of them were lucky to be in the right place and time with pre-existng wealth.

But you or I would never have developed something like Facebook anyway because building an empire on data mining and manipulating millions of people for greed would simply not occur to us.

They got lucky, but you also have to be a psychopath to do what they do, and amass that wealth without getting to a point where you just start giving it all back. When your entire family has everything a human could ever want, and will do so forever, how do you keep piling up money you'll never spend knowing people are suffering? How do you increase that suffering on purpose for more dollars that mean nothing to you?

Consider that anyone of them could use their money to become a hero of the human race, remembered forever. They could end world hunger, eradicate disease, cure cancer with that wealth. But even that isn't enough to overcome their need for more and more gold. These people are fucking monstrous.
 
Upvote
52 (56 / -4)
I'm curious if/how facebook intends to deal with the question of why you would go with them.

Of all the 'AI' outfits they seem like the one with the least obvious messaging in terms of why you would possibly want to touch their offering rather than one of the others.

You've got anthropic and openai that at least present themselves as being relatively application-agnostic purveyors to anyone who is buying. Then you've got outfits like salesforce and MS who emphasize that their bot is at least at parity with the other bots; and is already wired up for purposes of CRM and guzzling stuff from sharepoint, respectively.

But facebook? Zuck has some dystopian parasocial hellscape use cases for which facebook itself is the customer; but what's the external offering and why would external parties go with it?
 
Upvote
20 (20 / 0)
In fact, they're usually dumber. If you look at most billionaires the only thing separating them from you and me is that they had access to wealth. Because of that wealth safety net, they can try and fail repeatedly until they hit on something that works. The average person starting a business has only 1 shot, while billionaires usually start wealthy so can try multiple times to hit it big.
I read a good analogy for this once.

"It's like taking a shot at a carnival game. Most people can afford one shot. The rich just keep trying until they win, not matter how bad they are at it. The super wealthy own the carnival. The poor are working the carnival and don't get to play at all."
 
Upvote
24 (25 / -1)

motytrah

Ars Tribunus Militum
2,987
Subscriptor++
Out of all of the bullshit genAI boondoggles out there, I understand Meta’s the least. Does anyone really think a few percentage points in ad conversions is worth hundreds of billions? If not that, what? We truly live in the idiotic time of lemmings.
I don't think they understand it either. Buddy of mine works at a new meta datacenter. It's empty. And they won't have stuff for 6-12 months. The more he talks about it the more I get the feeling that they are just winging it.
 
Upvote
29 (29 / 0)
I read a good analogy for this once.

"It's like taking a shot at a carnival game. Most people can afford one shot. The rich just keep trying until they win, not matter how bad they are at it. The super wealthy own the carnival. The poor are working the carnival and don't get to play at all."
Exactly. It's true even for "self-made" billionaires like Taylor Swift who aren't terribly exploiting huge numbers of people. Her parents were wealthy enough to afford to move the entire family to Nashville to let her pursue a music career. Parent's can't afford to move to Nashville on the long shot chance their daughter gets picked and have to keep working the jobs they have? No Taylor Swift music star.
 
Upvote
25 (25 / 0)

murty

Ars Centurion
344
Subscriptor++
the-only-winning-move.gif
 
Upvote
12 (13 / -1)
They could show that they're serious in their attempt to catch up by firing the moron who wasted billions of dollars and a few years on trying to make the Metaverse becoming a thing. What? No? That isn't a sacking matter? But doing a job that can be done 60% as well by AI, until it fucks up, is? OK.
All the tech companies are wild about AGI. Will a computer program be able to achieve the intelligence of an average human?

but

Would you trust an average human with no sense of ethics and no regard for procedure to handle your user account security? See also George Carlin about the 'average person.'
 
Upvote
17 (17 / 0)

DrewW

Ars Tribunus Militum
2,046
Subscriptor++
Out of all of the bullshit genAI boondoggles out there, I understand Meta’s the least. Does anyone really think a few percentage points in ad conversions is worth hundreds of billions? If not that, what? We truly live in the idiotic time of lemmings.
It’s worse than just better ads. Meta insists on putting AI places it doesn’t belong. For example, I was trying to find an old text in WhatsApp and it truncated the previous messages and told me AI could summarize them for me. I don’t want or need an AI summary, I need the texting app to remember the text.
 
Upvote
10 (10 / 0)
Exactly. It's true even for "self-made" billionaires like Taylor Swift who aren't terribly exploiting huge numbers of people. Her parents were wealthy enough to afford to move the entire family to Nashville to let her pursue a music career. Parent's can't afford to move to Nashville on the long shot chance their daughter gets picked and have to keep working the jobs they have? No Taylor Swift music star.
Just becoming a highly-paid professional requires some serious luck on the opportunities handed to you and how you pursue them. Just for me to become a senior software developer, my parents needed to get me on ADHD treatment, put me into good school systems, and saved up a college fund for me; while I got lucky figuring out that I wanted to program in middle & high-school, turned out to be (relatively) really good at it, carried that through college, and got a really good job at the end of it.

Meanwhile, I have friends who have the problem solving skills and aptitude to do what I do if only they had the opportunity to study it like I did. I speak to people daily who have the same level of skill and talent that I have, but struggle to find a paying job to apply it to because their calling was animation or 3D modeling instead of software development.
 
Upvote
19 (20 / -1)
When, if ever, are they going to recoup these billions? Some existing models are light-years ahead of Meta without the stinky reputation. Look at OpenAI who lost so much to Anthropic because their execs were giving back their VC funds to MAGA PAC's instead of making good products. Like it or not, Claude Code is extremely powerful. Meta only beginning to make models, without the goodwill given to Anthropic, will cause them a huge delay to catch-up a remotely sufficient market share. In comparison, how is Threads doing? How many money did they put in this social media which seems barely used? When will it provide real returns? How is the real user-base, not the inflated one for accounts which were automatically activated for instance? Meta is for all accounts just a pile of burning cash, fuelled by legacy products throwing the dollars in the bonfire, until that source dwindles so much there won't be anything left.
 
Upvote
8 (8 / 0)

Kavinsky

Smack-Fu Master, in training
72
It's actually worse than that. Yes, most of them were lucky to be in the right place and time with pre-existng wealth.

But you or I would never have developed something like Facebook anyway because building an empire on data mining and manipulating millions of people for greed would simply not occur to us.

They got lucky, but you also have to be a psychopath to do what they do, and amass that wealth without getting to a point where you just start giving it all back. When your entire family has everything a human could ever want, and will do so forever, how do you keep piling up money you'll never spend knowing people are suffering? How do you increase that suffering on purpose for more dollars that mean nothing to you?

Consider that anyone of them could use their money to become a hero of the human race, remembered forever. They could end world hunger, eradicate disease, cure cancer with that wealth. But even that isn't enough to overcome their need for more and more gold. These people are fucking monstrous.
Tom from MySpace calling in from his yacht off a sunny beach.

Made product, sold it early for hundred of millions, retired, never hear from him any more.

The ideal Tech Bro.
 
Upvote
22 (22 / 0)

JoHBE

Ars Praefectus
4,396
Subscriptor++
I'm curious if/how facebook intends to deal with the question of why you would go with them.

Of all the 'AI' outfits they seem like the one with the least obvious messaging in terms of why you would possibly want to touch their offering rather than one of the others.

You've got anthropic and openai that at least present themselves as being relatively application-agnostic purveyors to anyone who is buying. Then you've got outfits like salesforce and MS who emphasize that their bot is at least at parity with the other bots; and is already wired up for purposes of CRM and guzzling stuff from sharepoint, respectively.

But facebook? Zuck has some dystopian parasocial hellscape use cases for which facebook itself is the customer; but what's the external offering and why would external parties go with it?
They might count on him being the least scrupulous to multiply their investment dollars ANY way possible?
 
Upvote
2 (2 / 0)

MagicDot

Ars Scholae Palatinae
1,124
Subscriptor
AI seems to me to be a long-term war of attrition, so I would put my money on the organizations that have essentially limitless capital. Zuckerberg is an idiot, but his cash cow product (Facebook) is fueled by narcissism, bragging, and whining which we now have sufficient empirical evidence of being limitless. Winning the "AI war" is about being able to keep your money furnace burning longer than anyone else.
 
Upvote
7 (7 / 0)

bigcheese

Ars Praetorian
586
Subscriptor
Out of all of the bullshit genAI boondoggles out there, I understand Meta’s the least. Does anyone really think a few percentage points in ad conversions is worth hundreds of billions? If not that, what? We truly live in the idiotic time of lemmings.
Metas ad revenue in 2025 was 196.2B. A five percent uptick over ten years is literally worth a hundred billion.
 
Upvote
6 (6 / 0)

vought1221

Ars Scholae Palatinae
869
Subscriptor++
It’s gonna be real hard to build this stuff when most of the talent that is left at Facebook and Meta is doing their darndest to get the hell out of there.

Apparently, employees are smart enough to stop touching the stove, even if Mark isn’t. Having Dan White on the board is just another piece of sand in an oyster that will be making pearls elsewhere by this time next year.

Seriously, I’ve talked to a couple of decade plus Meta veterans in the past few days. They are fucking done With the surveillance and shitty treatment. They are going to lose many of their best people over the next 12 months.
 
Upvote
5 (5 / 0)

vought1221

Ars Scholae Palatinae
869
Subscriptor++
Upvote
1 (1 / 0)