Metin načrt porabe za umetno inteligenco v vrednosti 135 milijard dolarjev: kaj v resnici kupuje (in tveganje mehurčka)

/Tehnologija/ Avtor

Povzetek:Meta pravi, da bi lahko porabila do135 milijard dolarjevletos – skoraj dvakrat več kot lani, v primerjavi z porabo, povezano z umetno inteligenco – večinoma za infrastrukturo, ki poganja umetno inteligenco. To ni le zgodba o »večjem proračunu«. Gre za strateški privzem zemljišč za računalništvo, talente in distribucijo v trenutku, ko vodilni na področju tehnologije in financ odkrito razpravljajo o tem, ali je razcvet umetne inteligence posledica…gospodarski mehurček.

Ključno vprašanje ni, ali bo umetna inteligenca pomembna (kar bo). Vprašanje je, ali lahko Meta velikanske kapitalske izdatke pretvori v trajno prednost izdelka in dobiček – ne da bi se ponovili pretekli cikli, ko je navdušenje preseglo vrnitev.

Zakaj je ta zgodba večja od "povečanja kapitalskih izdatkov"

Preprostejša različica te zgodbe je: »Meta bo porabila več za umetno inteligenco.« Pomembnejša različica je: Meta si poskuša odkupiti vodilni položaj na naslednjem nivoju vmesnika – priporočila, pomočniki in agenti, ki jih poganja umetna inteligenca – preden se tržna struktura ustali.

Zato je vredno ločiti, kaj jepotrjeno(številke, izjave) iz tega, kar jeimplicitno(strategija in pričakovani rezultati).

Kaj je Meta dejansko rekla (konkretna dejstva)

Iz poročila:

  • Meta pričakuje, da bo porabilado 135 milijard dolarjev (97 milijard funtov)letos, večinoma naInfrastruktura umetne inteligence.
  • To se primerja s približno72 milijard dolarjevlani.
  • V zadnjih treh letih je Meta porabila približno140 milijard dolarjevlov na razcvet umetne inteligence.
  • Zuckerberg je dejal, da pričakuje2026leto, ko bo umetna inteligenca »dramatično spremenila naš način dela«.
  • Metini stroški naraščajo hitreje kot prihodki (pritisk na marže).
  • Zuckerberg je namignil, da bo umetna inteligenca stisnila delo, ki je prej zahtevalo velike ekipe.
  • Meta je že odpustila na stotine delavcev (zlasti v podjetju Reality Labs).

Te točke uokvirjajo zgodbo: Meta podvoji prepričanje, da se umetna inteligenca premika iz funkcije v operativno plast tako za izdelke kot za interno delo.

Kam dejansko gre denar (in zakaj je tako drag)

Ko podjetje reče »infrastruktura umetne inteligence«, običajno misli na kup stvari, ki so energijsko lačne in kapitalsko intenzivne.

Preprost način razmišljanja: Meta ne kupuje "umetne inteligence". Kupujeprepustnost—zmožnost hitrejšega učenja večjih modelov in izvajanje sklepanja v velikem obsegu za milijarde dnevnih interakcij.

To zahteva:

1) Računalniška strojna oprema

  • Gruče GPU/pospeševalnikov za učenje in zagon modelov.
  • Visoka pasovna širina pomnilnika, hitre medsebojne povezave, shranjevanje.

2) Podatkovni centri

  • fizične stavbe, stojala, redundanca
  • dobava električne energije (pogosto dolgoročne pogodbe o električni energiji)
  • hladilni sistemi (velika inženirska omejitev)

3) Mreženje

Usposabljanje velikih modelov zahteva na tisoče čipov, ki delujejo kot en sam računalnik. To zahteva:

  • visokohitrostne tkanine
  • nizka latenca
  • skrbna topologija in zanesljivost

4) Orodja in modeliranje

  • podatkovnih cevovodov
  • varnostni/ocenjevalni pasovi
  • uvajanje in spremljanje

Zato se kapitalski izdatki za umetno inteligenco razlikujejo od »običajne« naložbe v programsko opremo: ne morete kar najeti inženirjev. Kupiti morate elektriko + silicij + nepremičnine.

Metina strateška stava: umetna inteligenca + distribucija je jarek

Meta je eno redkih podjetij z globalno distribucijo za potrošnike na več platformah:

  • Facebook
  • Instagram
  • WhatsApp
  • (in sorodna prizadevanja na področju strojne opreme/AR)

Če umetna inteligenca postane primarni vmesnik za to, kako ljudje odkrivajo vsebine, komunicirajo in ustvarjajo medije, je distribucija pomembna.

Metina implicitna strategija je:

  1. agresivno vlagati v izgradnjo zmogljivosti in kapacitete modela
  2. namestite ga na površine, kjer ljudje že preživljajo čas
  3. te izboljšave spremenite v:
    • boljša angažiranost
    • boljša uspešnost oglasov
    • novi izdelki (asistenti, agenti, kreativna orodja)

Že majhne izboljšave v učinkovitosti ciljanja oglasov ali ustvarjanju kreativnih vsebin se lahko seštejejo, saj je Meta-in oglaševalski posel tako velik.

Trditev, da »umetna inteligenca dramatično spreminja delo«: kaj bi lahko pomenila

Zuckerbergovi komentarji o projektih, ki se krčijo iz »velikih ekip« na »eno samo, zelo nadarjeno osebo«, kažejo na zelo specifično smer: umetna inteligenca kot multiplikator produktivnosti znotraj podjetja.

V praksi bi to lahko izgledalo takole:

  • Programski inženirji uporabljajo umetno inteligenco za hitrejše pisanje, refaktoriranje, testiranje in dokumentiranje kode
  • vodje izdelkov, ki uporabljajo umetno inteligenco za sintezo povratnih informacij, ustvarjanje eksperimentov in pripravo specifikacij
  • tržniki ustvarjajo različice in hitro ponavljajo

Vendar obstaja en zanka: orodja za produktivnost so neenakomerna. Ljudje, ki se jih naučijo dobro uporabljati, dobijo veliko večjo vrednost. To se ujema z Zuckerbergovo pripombo o »veliki razliki« med ljudmi, ki to počnejo dobro, in tistimi, ki ne.

Zakaj se odpuščanja pojavijo v istem pogovoru

Ko vodilni delavci govorijo o zmanjševanju produktivnosti, so odpuščanja v ozadju.

To ne pomeni nujno, da »umetna inteligenca nadomesti vse«. Pogosteje pomeni:

  • manj ljudi je potrebnih za rutinska opravila
  • Od ekip se pričakuje, da bodo z manj dobavile več
  • organizacije ponovno razvrstijo, katere vloge so strateške

Odpuščanja v podjetju Reality Labs še posebej nakazujejo, da Meta preusmerja proračun z dolgoročnejših stav (strojna oprema metaverza) na kratkoročnejšo infrastrukturo umetne inteligence in integracijo izdelkov umetne inteligence.

Tveganje za mehurčke: zakaj pametni ljudje tihi del vedno znova izgovarjajo na glas

Članek omenja več voditeljev, ki so izrazili zaskrbljenost glede mehurčkov in primerjali trenutek z dobo dot-com.

To je pomemben odtenek: »mehurček« ne pomeni, da je »umetna inteligenca ponarejena«. Običajno pomeni:

  • preveč kapitala se zanaša na premalo očitno dobičkonosnih aplikacij
  • Mnoga podjetja ne bodo preživela pretresa
  • Zmagovalci infrastrukture in zmagovalci distribucije pridobijo največ vrednosti

Izvršni direktor Cisca je opozoril, da bodo zmagovalci sicer prišli, a da bo na poti "pokol". To je realističen opis tehnoloških prehodov.

Še ena lekcija o dot-com: med mehurčkom so podjetja zgradila pravo infrastrukturo (optična vlakna, podatkovne centre, omrežja). Velik del zgodnje vrednosti lastniškega kapitala je izhlapel – vendar je infrastruktura ostala in kasneje omogočila sodobno internetno gospodarstvo. Današnja gradnja umetne inteligence bi lahko sledila istemu vzorcu: boleče pretresanje za nekatera podjetja, a dolgotrajna zmogljivost, ki postane temeljna.

Metain profil tveganja: štirje načini, kako lahko gre kaj narobe

1) Kapitalski izdatki brez trajne diferenciacije izdelkov

Če konkurenti hitro uskladijo zmogljivosti, postane poraba višja – draga, a ne diferencirajoča.

2) Podcenjevanje obratovalnih stroškov

Nakup strojne opreme je šele začetek. Usposabljanje modelov in sklepanje:

  • elektrika
  • zmogljivost mreženja
  • čas inženiringa za oceno in varnost

Če se operativni stroški povečujejo hitreje kot prihodki, postane »prednost umetne inteligence« ovira za maržo.

3) Pritisk na marže in potrpežljivost vlagateljev

Meta si lahko privošči veliko porabo, dokler njen osrednji oglasni mehanizem ostane močan. Če pa se makroekonomski pogoji ali angažiranost spremenijo, bodo vlagatelji tveganje prevrednotili.

4) Regulativna vprašanja in vprašanja zaupanja

Razvrščanje in generiranje na podlagi umetne inteligence vzbuja pomisleke glede:

  • širjenje dezinformacij
  • globoke ponaredke in goljufije
  • napake pri moderiranju vsebine
  • meje zasebnosti v aplikacijah za sporočanje

Če funkcije umetne inteligence ustvarijo več škode kot koristi, lahko regulatorji zaostrijo omejitve in s tem zmanjšajo rast.

Kako izgleda uspeh (signali, ki jih je vredno spremljati)

Če želite presoditi, ali Meta-ina poraba za umetno inteligenco deluje, prezrite sporočila za javnost in poiščite merljive signale.

Koristno uokvirjanje: Meta potrebuje umetno inteligenco za izboljšanje obehprihodek na uporabnika,stroški na enoto proizvodnje, ali idealno oboje. Če teh ne vidite skozi čas, investicijska teza oslabi.

1) Izboljšave izdelkov, ki se ohranijo

  • boljša priporočila, ki podaljšajo porabljen čas brez povečanja pritožb
  • ustvarjalna orodja, ki resnično zmanjšujejo trenje za oglaševalce in ustvarjalce

2) Poslovna uspešnost

  • oblikovanje cen oglasov in kakovost konverzij
  • stroški na rezultat za oglaševalce
  • ali se rast prihodkov pospeši glede na rast stroškov

3) Zmogljivost modela in hitrost uvajanja

  • kako hitro se novi modeli uvajajo v aplikacije
  • ali postanejo »agenti« uporabni v običajnih delovnih procesih (ne le v demonstracijah)

4) Varnost in zaupanje

  • kako dobro Meta vsebuje zlorabe (prevare, lažno predstavljanje, sintetični mediji)
  • preglednost glede vsebin, ustvarjenih z umetno inteligenco

Praktični vodnik za bralce: kaj verjeti in kaj obravnavati kot trženje

Obvestila o umetni inteligenci pogosto mešajo trdne inženirske realnosti z narativnim okvirjem. Koristen kontrolni seznam:

  • Če gre začipi, napajanje, podatkovni centri, je resnično in merljivo.
  • Če gre zaagenti, ki spreminjajo delo, vprašajte, kateri delovni procesi so danes dejansko izboljšani.
  • Če gre zaprihranki stroškov, se vprašajte, ali se prihranki kažejo v maržah ali zgolj financirajo večjo rast.

Bistvo

Meta troši kot podjetje, ki verjame, da je umetna inteligenca naslednji premik platforme – in da je prava poteza zavarovati računalništvo in uvesti umetno inteligenco povsod, kjer so njeni uporabniki že.

Obseg je bistvo zgodbe: Meta se odloča za konkurenco na področju infrastrukture in distribucije, ne le na področju pametnih namigov. To je vrsta zaveze, ki lahko ustvari jarek – ali zelo drago napako.

Prednosti so resnične: boljši izdelki, boljši oglasi, novi pomočniki in ustvarjalna orodja. Slabosti so prav tako resnične: zniževanje marž, prenatrpano področje umetne inteligence in tveganje, da bodo težave s regulacijo in zaupanjem zmanjšale donose.

Takole izgleda prehod platforme v realnem času: ogromne naložbe v infrastrukturo, glasen skepticizem in tekma za dokazovanje, da se poraba spremeni v trajno prednost.

Če Meta lahko pokaže trajne izboljšave v učinkovitosti oglasov in pristnosti izdelkov, hkrati pa ohrani zaupanje in varnost pod nadzorom, bodo kapitalski izdatki videti kot predvidevanje. Če ne, tvega, da bo postala odmeven primer, kako enostavno je v ciklu navdušenja porabiti preveč.

Viri

Document Title
Meta may spend up to $135bn on AI infrastructure: where the money goes and what investors should watch
Meta may spend up to $135bn on AI infrastructure this year. Here’s what that buys (chips, data centres, networking), the strategy behind it, and the bubble risk investors debate.
Title Attribute
oEmbed (JSON)
oEmbed (XML)
JSON
View all posts by Admin
Tesla ends Model S/X as it pivots to robots and AI — what’s real vs hype
UK’s free AI training push: why ‘prompting’ is the easy part
Page Content
Meta may spend up to $135bn on AI infrastructure: where the money goes and what investors should watch
Nature
Climate
Meta’s $135bn AI spending plan: what it’s really buying (and the bubble risk)
/
Technology
/ By
Admin
Summary:
Meta says it could spend up to
$135bn
this year—nearly double last year’s AI-related spend—mostly on infrastructure that powers artificial intelligence. This is not just a “bigger budget” story. It’s a strategic land grab for compute, talent, and distribution at a moment when leaders across tech and finance are openly debating whether the AI boom is an
economic bubble
.
The key question isn’t whether AI will matter (it will). The question is whether Meta can translate giant capex into durable product advantage and profit—without repeating past cycles where enthusiasm outran returns.
Why this story is bigger than “capex goes up”
The easy version of this story is “Meta will spend more on AI.” The more important version is: Meta is trying to buy its way into a leadership position in the next interface layer—AI-powered recommendations, assistants, and agents—before the market structure settles.
That’s why it’s worth separating what’s
confirmed
(numbers, statements) from what’s
implied
(strategy and expected outcomes).
What Meta actually said (the concrete facts)
From the reporting:
Meta expects to spend
up to $135bn (£97bn)
this year, mostly on
AI infrastructure
That compares with roughly
$72bn
last year.
Over the last three years, Meta has spent about
$140bn
chasing the AI boom.
Zuckerberg said he expects
2026
to be the year AI “dramatically changes the way we work.”
Meta’s expenses have been rising faster than revenue (pressure on margins).
Zuckerberg hinted that AI will compress work that used to require big teams.
Meta has already laid off hundreds of workers (notably in Reality Labs).
Those points frame the story: Meta is doubling down on the belief that AI is shifting from a feature to an operating layer for both products and internal work.
Where the money actually goes (and why it’s so expensive)
When a company says “AI infrastructure,” it usually means a stack of things that are power-hungry and capital-intensive.
One simple way to think about it: Meta isn’t buying “AI.” It’s buying
throughput
—the ability to train bigger models faster, and run inference at scale for billions of daily interactions.
That requires:
1) Compute hardware
GPU/accelerator clusters to train and run models.
High memory bandwidth, fast interconnects, storage.
2) Data centres
physical buildings, racks, redundancy
power delivery (often long-term power contracts)
cooling systems (a major engineering constraint)
3) Networking
Training large models requires thousands of chips acting like one computer. That demands:
high-speed fabrics
low latency
careful topology and reliability
4) Tooling and model operations
data pipelines
safety/evaluation harnesses
deployment and monitoring
This is why AI capex has a different shape from a “normal” software investment: you can’t just hire engineers. You must buy electricity + silicon + real estate.
Meta’s strategic bet: AI + distribution is a moat
Meta is one of the few companies with global consumer distribution across multiple surfaces:
Facebook
Instagram
WhatsApp
(and adjacent efforts in hardware/AR)
If AI becomes a primary interface for how people discover content, communicate, and create media, distribution matters.
Meta’s implicit strategy is:
invest aggressively to build model capability and capacity
deploy it across the surfaces where people already spend time
turn those improvements into:
better engagement
better ad performance
new products (assistants, agents, creative tools)
Even small improvements in ad targeting efficiency or creative generation can compound, because Meta’s ad business is so large.
The “AI dramatically changes work” claim: what it could mean
Zuckerberg’s comments about projects shrinking from “big teams” to “a single, very talented person” signals a very specific direction: AI as a productivity multiplier inside the company.
In practice, that could look like:
software engineers using AI to write, refactor, test, and document code faster
product managers using AI to synthesize feedback, generate experiments, draft specs
marketers generating variants and iterating quickly
But there’s a catch: productivity tools are uneven. People who learn to use them well get much more value. That aligns with Zuckerberg’s comment about a “big delta” between people who do it well and those who don’t.
Why layoffs show up in the same conversation
When executives talk about productivity compression, layoffs are the shadow topic.
It doesn’t necessarily mean “AI replaces everyone.” More often it means:
fewer people needed for routine tasks
teams are expected to ship more with less
organisations re-rank which roles are strategic
Reality Labs layoffs in particular hint that Meta is shifting budget away from longer-horizon bets (metaverse hardware) toward nearer-term AI infrastructure and AI product integration.
Bubble risk: why smart people keep saying the quiet part out loud
The article notes multiple leaders raising bubble concerns, comparing the moment to the dot-com era.
This is an important nuance: “bubble” doesn’t mean “AI is fake.” It usually means:
too much capital is chasing too few clearly profitable applications
many companies will not survive the shakeout
infrastructure winners and distribution winners capture most value
Cisco’s CEO is quoted warning that winners will emerge but there will be “carnage along the way.” That’s a realistic description of technology transitions.
One more dot-com lesson: during the bubble, companies built real infrastructure (fibre, data centres, networks). Much of the early equity value evaporated—but the infrastructure remained and later enabled the modern internet economy. Today’s AI buildout could follow the same pattern: painful shakeout for some firms, but long-lived capacity that becomes foundational.
Meta’s risk profile: four ways this can go wrong
1) Capex without durable product differentiation
If competitors match capabilities quickly, the spend becomes table stakes—expensive, but not differentiating.
2) Underestimating operating costs
Buying hardware is only the start. Model training and inference burn:
electricity
networking capacity
engineering time for evaluation and safety
If operating costs scale faster than revenue lift, the “AI advantage” becomes a margin drag.
3) Margin pressure and investor patience
Meta can afford large spend as long as its core ad engine remains strong. But if macro conditions or engagement shift, investors will reprice the risk.
4) Regulatory and trust issues
AI-driven ranking and generation raises concerns about:
misinformation amplification
deepfakes and fraud
content moderation errors
privacy boundaries in messaging apps
If AI features create more harm than value, regulators may tighten constraints, reducing upside.
What success looks like (signals worth watching)
If you want to judge whether Meta’s AI spending is working, ignore press releases and look for measurable signals.
A useful framing: Meta needs AI to improve either
revenue per user
,
cost per unit of output
, or ideally both. If you can’t see those showing up over time, the investment thesis weakens.
1) Product improvements that stick
better recommendations that increase time spent without increasing complaints
creative tools that genuinely reduce friction for advertisers and creators
2) Business performance
ad pricing and conversion quality
cost per outcome for advertisers
whether revenue growth accelerates relative to expense growth
3) Model capability and deployment pace
how quickly new models are deployed across apps
whether “agents” become useful in normal workflows (not just demos)
4) Safety and trust
how well Meta contains abuse (scams, impersonation, synthetic media)
transparency about AI-generated content
A practical reader’s guide: what to believe and what to treat as marketing
AI announcements often mix solid engineering realities with narrative framing. A useful checklist:
If it’s about
chips, power, data centres
, it’s real and measurable.
agents changing work
, ask what workflows are actually improved today.
cost savings
, ask whether savings show up in margins or just fund more growth.
Bottom line
Meta is spending like a company that believes AI is the next platform shift—and that the right move is to secure compute and deploy AI everywhere its users already are.
The scale is the story: Meta is choosing to compete on infrastructure and distribution, not just on clever prompts. That’s the kind of commitment that can create a moat—or a very expensive mistake.
The upside is real: better products, better ads, new assistants and creative tools. The downside is also real: margin compression, a crowded AI field, and the risk that regulation and trust problems blunt the returns.
This is what a platform transition looks like in real time: huge infrastructure investment, loud skepticism, and a race to prove that the spend turns into durable advantage.
If Meta can show sustained improvements in ad performance and product stickiness while keeping trust and safety under control, the capex will look like foresight. If not, it risks becoming a high-profile example of how easy it is to overspend in a hype cycle.
Sources
BBC News (Technology):
https://www.bbc.com/news/articles/cn8jkyk78gno?at_medium=RSS&at_campaign=rss
BBC News (Technology) (related context on bubble concerns):
https://www.bbc.com/news/articles/cr57p2ve8glo
←
Previous Post
Next Post
→
oEmbed (JSON)
oEmbed (XML)
JSON
View all posts by Admin
Tesla ends Model S/X as it pivots to robots and AI — what’s real vs hype
UK’s free AI training push: why ‘prompting’ is the easy part
Meta may spend up to $135bn on AI infrastructure this year. Here’s what that buys (chips, data centres, networking), the strategy behind it, and the bubble risk investors debate.
Document Title
Page not found - Florin.blog
Image Alt
Florin.blog
Title Attribute
Florin.blog » Feed
RSD
Skip to content
Placeholder Attribute
Search...
Page Content
Page not found - Florin.blog
Skip to content
Home
Blog
Garden Decor
Indoor
Main Menu
This page doesn't seem to exist.
It looks like the link pointing here was faulty. Maybe try searching?
Search for:
Search
Quick Links
Outdoors
About
Contact
Explore
Bestsellers
Hot deals
Best of The Year
Featured
Gift Cards
Help
Privacy Policy
Disclaimer
: As an Amazon Associate, we earn from qualifying purchases — at no extra cost to you.
Florin.blog
Florin.blog » Feed
RSD
Search...
l Slovenščina