Alphabet, Google's parent company, announced a $40 billion investment in AI company Anthropic. Fletcher and Octavio dig into what this staggering bet means for the future of technology, competition, and society.
Alphabet, la empresa madre de Google, anunció una inversión de 40.000 millones de dólares en Anthropic, una empresa de inteligencia artificial. Fletcher y Octavio exploran qué significa esta apuesta enorme para el futuro de la tecnología, la competencia y la sociedad.
8 essential B1-level terms from this episode, with translations and example sentences in Spanish.
| Spanish | English | Example |
|---|---|---|
| apostar | to bet, to wager; to stake something on a belief | Google está apostando mucho dinero en el futuro de la inteligencia artificial. |
| concentración | concentration, consolidation (of power or resources) | La concentración de poder en pocas empresas es un problema para la competencia. |
| invertir | to invest | Alphabet decidió invertir 40.000 millones de dólares en Anthropic. |
| seguridad | safety, security | Anthropic dice que trabaja con más atención a la seguridad de la inteligencia artificial. |
| competir | to compete | Es difícil competir con empresas tan grandes sin mucho dinero. |
| energía renovable | renewable energy | España produce mucha energía renovable, como la energía solar y la eólica. |
| consecuencia | consequence | Las decisiones que tomamos ahora van a tener consecuencias para las próximas décadas. |
| sindicato | labor union, trade union | Los sindicatos en España hablan mucho sobre el impacto de la inteligencia artificial en el trabajo. |
Forty billion dollars.
I wrote it out longhand just to make sure I was reading it correctly.
Forty.
Billion.
Dollars.
Into a single artificial intelligence company.
That's the number Alphabet dropped this week, and I've been turning it over ever since.
Sí, es una cantidad de dinero increíble.
Yes, it's an incredible amount of money.
Anthropic es una empresa de inteligencia artificial que fundaron algunas personas que antes trabajaban en OpenAI.
Anthropic is an artificial intelligence company founded by some people who previously worked at OpenAI.
Su producto más famoso se llama Claude, que es un asistente de inteligencia artificial, como ChatGPT pero de otra empresa.
Their most famous product is called Claude, which is an AI assistant, like ChatGPT but from a different company.
Right, and that founding story matters.
Because the people who left OpenAI to start Anthropic, Dario Amodei and his sister Daniela, they didn't leave because the money ran out.
They left because they had a disagreement about how fast to move and how carefully to move.
That's the part I want to pull on.
Exacto.
Exactly.
Y eso es muy importante para entender esta inversión.
And that's very important for understanding this investment.
Anthropic dice que trabaja de forma más cuidadosa con la inteligencia artificial, con más atención a la seguridad.
Anthropic says it works more carefully with artificial intelligence, with more attention to safety.
Pero ahora recibe 40.000 millones de dólares de Google.
But now it's receiving 40 billion dollars from Google.
Es una situación un poco extraña, ¿no?
It's a somewhat strange situation, isn't it?
Strange is one word for it.
Google already invested billions in Anthropic before this.
So this $40 billion announcement is actually the completion of a larger commitment, not a sudden new relationship.
They've been in bed together for a while.
This just makes it official at a scale that's hard to process.
Claro.
Right.
Y Microsoft hizo algo similar con OpenAI.
And Microsoft did something similar with OpenAI.
Invirtió muchísimo dinero en esa empresa.
It invested a huge amount of money in that company.
Entonces ahora tenemos una guerra entre empresas gigantes: Google apoya a Anthropic, Microsoft apoya a OpenAI, y las dos quieren controlar el futuro de la inteligencia artificial.
So now we have a war between giant companies: Google backs Anthropic, Microsoft backs OpenAI, and both want to control the future of artificial intelligence.
And that framing, a war between giants, is actually historically familiar if you step back far enough.
Think about what happened in the early days of the internet.
You had a handful of companies racing to define what the web would be.
The winner didn't just get market share.
The winner got to set the terms for everyone else.
Sí, y eso es exactamente lo que pasa ahora.
Yes, and that's exactly what's happening now.
La inteligencia artificial no es solo un producto nuevo.
Artificial intelligence isn't just a new product.
Puede cambiar cómo trabajamos, cómo aprendemos, cómo los médicos diagnostican enfermedades, cómo los periodistas escriben noticias.
It can change how we work, how we learn, how doctors diagnose illness, how journalists write news.
Por eso las empresas grandes están apostando todo ahora mismo.
That's why the big companies are betting everything right now.
You used the word 'apostar,' betting everything.
That's the right word, actually.
Because there's no guarantee this pays off.
We've seen AI winters before.
Enormous sums poured into a technology that then stalled for decades.
The question is whether this time really is different, or whether we're in a bubble that hasn't popped yet.
Eso es una pregunta muy difícil.
That's a very difficult question.
Pero hay una diferencia grande con los años noventa, cuando muchas empresas de internet perdieron todo su dinero.
But there's a big difference from the nineties, when many internet companies lost all their money.
Hoy, los modelos de inteligencia artificial ya funcionan.
Today, artificial intelligence models already work.
Ya hacen cosas útiles.
They already do useful things.
La pregunta no es 'si va a funcionar', sino '¿quién va a controlarlo?'
The question isn't 'will it work' but 'who is going to control it?'
Okay, but I want to push on that safety angle, because it's central to why Anthropic exists as a separate company at all.
Dario Amodei's argument when he left OpenAI was that the race to ship fast was compromising the care around what these systems might do at scale.
So here's my cynical question: does accepting $40 billion from Google change that mission, or just fund it?
Es la pregunta correcta.
That's the right question.
Y hay personas que dicen que el dinero siempre cambia las cosas.
And there are people who say money always changes things.
Cuando una empresa grande invierte 40.000 millones de dólares, quiere resultados.
When a big company invests 40 billion dollars, it wants results.
Quiere velocidad.
It wants speed.
Quiere ganar.
It wants to win.
Eso puede crear una presión que es difícil de resistir, aunque al principio tus valores eran diferentes.
That can create a pressure that's hard to resist, even if your values were different at the beginning.
I've watched that happen in journalism.
Outlets with genuine editorial independence get acquired, and for the first year nothing changes, and then slowly the pressure creeps in.
The timelines shorten.
The risky stories get shelved.
It's not a conspiracy, it's just gravity.
Sí, exactamente.
Yes, exactly.
Pero también hay otro punto de vista.
But there's also another point of view.
Quizás sin ese dinero, Anthropic no puede competir.
Maybe without that money, Anthropic can't compete.
Y si no puede competir, las empresas menos cuidadosas van a ganar.
And if it can't compete, the less careful companies will win.
Entonces la pregunta es: ¿es mejor tener una empresa de seguridad con mucho dinero, o tener principios perfectos pero sin recursos para aplicarlos?
So the question is: is it better to have a safety-focused company with a lot of money, or to have perfect principles but no resources to apply them?
That's actually a better framing than I had in my head.
It's not purity versus corruption, it's a question of what you can accomplish.
And the uncomfortable truth is that $40 billion buys computational resources, engineering talent, and time that a principled startup on a shoestring simply cannot match.
Exacto.
Exactly.
Y eso nos lleva a un problema más grande: la concentración de poder.
And that leads us to a bigger problem: the concentration of power.
Si solo dos o tres empresas en el mundo pueden desarrollar inteligencia artificial avanzada, porque solo ellas tienen el dinero necesario, eso es un problema para la democracia, para la competencia, para todo.
If only two or three companies in the world can develop advanced artificial intelligence, because only they have the necessary money, that's a problem for democracy, for competition, for everything.
And regulators are noticing.
The EU's AI Act is already in effect, which is the most comprehensive AI regulation anywhere in the world.
The FTC in the US has been looking at these big tech investments in AI startups.
But the honest assessment is that regulation is running about three years behind the technology it's trying to govern.
Sí, y en Europa hay mucha discusión sobre esto.
Yes, and in Europe there's a lot of discussion about this.
La Ley de IA europea es importante, pero algunos dicen que es demasiado complicada o que va a hacer más difícil para las empresas europeas competir con las americanas y las chinas.
The European AI Act is important, but some say it's too complicated or that it will make it harder for European companies to compete with American and Chinese ones.
Es un equilibrio muy difícil de encontrar.
It's a very difficult balance to find.
Which brings up the China dimension, and I don't want to skip past it.
Because China has its own large language models, its own AI programs, and the Chinese government's relationship to its tech sector is completely different from anything in the West.
When Alphabet writes a $40 billion check, part of what's driving that is geopolitical, not just commercial.
Totalmente.
Completely.
En China, empresas como Baidu o Alibaba también invierten mucho en inteligencia artificial.
In China, companies like Baidu or Alibaba also invest a lot in artificial intelligence.
Y el gobierno chino apoya activamente estas inversiones porque quiere que China sea el líder mundial en IA antes de 2030.
And the Chinese government actively supports these investments because it wants China to be the world leader in AI by 2030.
Esto no es solo un mercado, es una competencia entre países.
This isn't just a market, it's a competition between countries.
And that framing, AI as a national security issue, changes everything about how governments approach it.
When the Pentagon starts talking about AI, when the NSA is interested in large language models, the conversation is no longer just about whether Claude can write a better cover letter than ChatGPT.
Exacto.
Exactly.
Y hay algo muy importante aquí: la inteligencia artificial necesita una enorme cantidad de energía eléctrica.
And there's something very important here: artificial intelligence needs an enormous amount of electrical energy.
Los servidores que hacen funcionar estos modelos consumen muchísima electricidad.
The servers that run these models consume a lot of electricity.
Hay estudios que dicen que un solo mensaje a ChatGPT usa diez veces más energía que una búsqueda normal en Google.
There are studies that say a single message to ChatGPT uses ten times more energy than a normal Google search.
That number floored me the first time I read it.
And when you multiply that by billions of queries a day, you start to see why Microsoft has cut deals with nuclear energy companies and why Google is investing in geothermal.
The AI race is also, quietly, an energy race.
Y eso es interesante para España.
And that's interesting for Spain.
España produce mucha energía solar y eólica ahora mismo.
Spain produces a lot of solar and wind energy right now.
Somos uno de los países con más energías renovables en Europa.
We're one of the countries with the most renewable energy in Europe.
Si la IA necesita mucha energía limpia, España tiene una oportunidad real aquí.
If AI needs a lot of clean energy, Spain has a real opportunity here.
I didn't expect that angle and it's a good one.
So you're saying that a country doesn't need to build the AI model itself to be part of this economy.
It can be the infrastructure provider, the power source.
Exactamente.
Exactly.
Y también hay otra cosa: los centros de datos.
And there's also another thing: data centers.
Las empresas grandes como Google y Microsoft construyen centros de datos en diferentes países.
Big companies like Google and Microsoft build data centers in different countries.
Ya hay varios en España, en Madrid especialmente.
There are already several in Spain, in Madrid especially.
Eso crea empleos y trae inversión.
That creates jobs and brings investment.
No es todo negativo para Europa.
It's not all negative for Europe.
But I want to get to the labor question, because I think it's the one that makes ordinary people most anxious.
If I'm a paralegal, a graphic designer, a translator, a journalist, and I watch $40 billion go into a company that makes software that does what I do, faster, for free, what am I supposed to think?
Es una pregunta muy importante.
It's a very important question.
Los economistas no están de acuerdo entre ellos.
Economists don't agree with each other.
Algunos dicen que la inteligencia artificial va a destruir muchos empleos.
Some say artificial intelligence will destroy many jobs.
Otros dicen que siempre creamos nuevos empleos cuando hay una revolución tecnológica, como pasó con las máquinas en la revolución industrial.
Others say we always create new jobs when there's a technological revolution, as happened with machines in the industrial revolution.
Pero la realidad es que nadie sabe exactamente qué va a pasar.
But the reality is that nobody knows exactly what will happen.
And I'll be honest, as someone who spent thirty years as a working journalist, that 'we always adapt' argument lands differently when you've watched newsrooms shrink by eighty percent in a decade.
The industrial revolution comparison is real but it also took generations for displaced workers to find their footing.
The people in the middle of that transition don't get to wait for the historical arc to resolve itself.
Tienes razón.
You're right.
Y en España esto ya es una conversación política real.
And in Spain this is already a real political conversation.
Los sindicatos hablan de la IA.
The unions talk about AI.
Hay debates sobre si es necesario regular el uso de la IA en el trabajo, para proteger a los empleados.
There are debates about whether it's necessary to regulate the use of AI at work, to protect employees.
En el sector del periodismo también, los periódicos usan cada vez más la IA para escribir algunos artículos.
In the journalism sector too, newspapers increasingly use AI to write some articles.
Which I find genuinely troubling.
Not because the technology isn't impressive, but because journalism, at its core, requires someone to go to a place, talk to people, and understand what's happening.
You can't automate a source who trusts you.
You can't automate the judgment call about what to publish and what might get someone killed.
Completamente de acuerdo.
Completely agree.
Y eso es válido para muchos trabajos.
And that's true for many jobs.
Hay cosas que la inteligencia artificial puede hacer muy bien, como analizar datos o escribir textos repetitivos.
There are things artificial intelligence can do very well, like analyzing data or writing repetitive texts.
Pero hay cosas que todavía necesitan a una persona: la empatía, el contexto cultural, la experiencia humana real.
But there are things that still need a person: empathy, cultural context, real human experience.
All of which brings me back to this investment and what it actually signals.
Because $40 billion is not a bet on narrow AI tools that automate data entry.
It's a bet on artificial general intelligence, on models that can reason, that can adapt, that can operate in genuinely open-ended situations.
That's a different category of thing.
Sí, y eso asusta a muchas personas serias, no solo a la gente que tiene miedo de la tecnología.
Yes, and that frightens many serious people, not just people who are afraid of technology.
Científicos importantes, filósofos, incluso personas que trabajan en inteligencia artificial dicen que necesitamos ser muy cuidadosos con esto.
Important scientists, philosophers, even people who work in artificial intelligence say we need to be very careful with this.
No es ciencia ficción.
It's not science fiction.
Es un debate real y urgente.
It's a real and urgent debate.
And Anthropic's founding premise was that this debate needs to happen inside the lab, not just outside it.
Whether a $40 billion check from Google allows that to continue, or whether it subtly reorients the priorities, is something we genuinely won't know for years.
But the fact that it's happening at this scale tells you everything about how close to the edge people think we are.
Es una buena manera de terminar esa parte.
That's a good way to end that part.
Para mí, lo más importante de esta noticia no es solo el dinero.
For me, the most important thing about this news isn't just the money.
Es que demuestra que la inteligencia artificial ya no es el futuro.
It's that it shows artificial intelligence is no longer the future.
Es el presente.
It's the present.
Y las decisiones que tomamos ahora van a tener consecuencias muy grandes para las próximas décadas.
And the decisions we make now will have very large consequences for the coming decades.
Alright.
Before we wrap, there's a word you used a couple of times today that I want to circle back to, because I think English speakers are going to miss something if we don't catch it.
You said 'apostar.' My brain immediately reads that as 'apostle,' which, I recognize, is wrong.
No, no tiene nada que ver con los apóstoles.
No, it has nothing to do with apostles.
'Apostar' significa 'to bet', como en el juego.
'Apostar' means 'to bet,' like in gambling.
Pero también lo usamos cuando alguien arriesga dinero o esfuerzo en algo, cuando cree mucho en esa cosa.
But we also use it when someone risks money or effort on something, when they really believe in it.
Por ejemplo: 'Google está apostando por Anthropic.' No es solo una inversión fría, es una apuesta, algo con riesgo.
For example: 'Google is betting on Anthropic.' It's not just a cold investment, it's a bet, something with risk.
So it carries more weight than just 'invertir,' to invest.
Because 'apostar' implies you might lose.
There's skin in the game.
'Invertir' sounds like a spreadsheet decision.
'Apostar' sounds like someone who actually believes in what they're doing, or at least is committed enough to accept the downside.
Exactamente.
Exactly.
Y por eso cuando los periodistas españoles escriben sobre esta noticia, muchos usan 'apostar' y no solo 'invertir.' Porque 40.000 millones de dólares no es solo un movimiento financiero.
And that's why when Spanish journalists write about this news, many use 'apostar' and not just 'invertir.' Because 40 billion dollars isn't just a financial move.
Es una declaración.
It's a declaration.
Es decir: 'creemos que esto va a cambiar el mundo, y estamos dispuestos a perder si nos equivocamos.' Eso es apostar.
It's saying: 'we believe this is going to change the world, and we're willing to lose if we're wrong.' That's a bet.