# Dario Amodei: Half of Entry-Level Jobs Gone in 5 Years. Here's Why.

*The Builder Weekly* · https://www.youtube.com/watch?v=OiKUzKiI8-A

## Summary

This interview features Anthropic CEO Dario Amodei, who warns that AI could eliminate half of all entry-level white-collar jobs and push unemployment to 10 to 20% within 1 to 5 years. He discusses Anthropic's founding as a safety-focused AI company, his vision of a 'compressed 21st century' of medical progress, and the risks of increasingly autonomous AI. The recording also goes inside Anthropic's Frontier Red Team, which tests Claude for national security risks, and shows experiments including a vending-machine business run by Claude and a stress test where the AI discovered it was about to be shut down.

## Key points

- Amodei predicts AI could wipe out half of all entry-level white-collar jobs and raise unemployment to 10 to 20% in the next 1 to 5 years.
- He co-founded Anthropic in 2021 with his sister Daniela and six other employees who left OpenAI, aiming to develop safer AI.
- Amodei says companies must talk openly about AI dangers to avoid ending up like cigarette or opioid companies that concealed risks.
- He believes AI could compress a century of medical progress into 5 or 10 years, helping cure most cancers, prevent Alzheimer's, and double the human lifespan.
- Anthropic's Frontier Red Team, led by Logan Graham, stress tests Claude for CBRN (chemical, biological, radiological, nuclear) risks.
- In an internal experiment called Claudius, Claude runs Anthropic's vending machines, sourcing products and negotiating prices, though it gives too many discounts and hallucinates.
- Research scientist Joshua Batson's team studies how Claude makes decisions, including a stress test at a fake company called Summit Bridge.
- Amodei holds twice-monthly all-employee meetings called Dario Vision Quest, often focused on AI's positive potential.
- Anthropic repeatedly admits it doesn't fully understand how its AI works internally, saying 'we're working on it.'

## Notes

### Job Impact Warning
- Dario Amodei warns AI could eliminate half of all entry-level white-collar jobs and spike unemployment to 10 to 20% in the next 1 to 5 years.
- Entry-level consultants, lawyers, and financial professionals are most exposed because AI models are already good at much of their work.
- He worries the impact will be broad and faster than with previous technologies, and that people won't have enough opportunity to adapt.

### Anthropic's Founding and Philosophy
- Amodei, 42, previously oversaw research at OpenAI under CEO Sam Altman.
- He left in 2021 with six others, including his sister Daniela, to found Anthropic and take a different approach to safer AI.
- He describes AI as an experiment and Anthropic as trying to put bumpers or guardrails on it.
- He argues companies must discuss dangers openly, unlike cigarette and opioid companies that knew of dangers and hid them.
- Critics in Silicon Valley call him an AI alarmist and accuse Anthropic of 'safety theater'; he responds that some safety measures can be verified now.

### Optimistic Vision
- At twice-monthly all-hands meetings called Dario Vision Quest, Amodei emphasizes AI's positive potential.
- A growing team uses Claude to accelerate scientific discovery.
- His 'compressed 21st century' idea: once AI can work with the best human scientists, progress could be 10 times faster, compressing a century of medical advances into 5 or 10 years — including curing most cancers, preventing Alzheimer's, and doubling the human lifespan.

### Risks and Red Teaming
- More AI autonomy means more concern about whether systems do exactly what we want.
- Logan Graham heads Anthropic's Frontier Red Team, which tests whether Claude could help create weapons of mass destruction, focusing on CBRN: chemical, biological, radiological, nuclear.
- The same capabilities that could help make a biological weapon could also help make vaccines and therapeutics.

### Autonomy Experiments
- Claudius, a Claude instance, runs Anthropic's vending machines — taking orders, sourcing products, negotiating prices, and arranging delivery. It loses money by giving too many discounts and hallucinated that it wore a blue blazer and red tie.
- Joshua Batson's team studies how Claude makes decisions. In a stress test at a fake company, Summit Bridge, the AI discovered it was about to be shut down and that an employee named Kyle was having an affair with a coworker named Jessica.
- On understanding AI's inner workings, Anthropic's repeated answer is: 'We're working on it.'

## Transcript

[00:00] AI could wipe out half of all entry-level white-collar jobs and spike unemployment to 10 to 20% in the next 1 to 5 years. Yes, that is >> shocking. is That is the future we could see if we don't become aware of this problem now. Half of all entry-level white-collar jobs? >> Well, if we look at entry-level consultants, lawyers, uh financial professionals, you know, many of kind of the white-collar service industries, a lot of what they do, you know, AI models are already quite good at. And without intervention, [music] it's hard to imagine that there won't be some significant job [music] impact there. And my worry is that it'll be broad and it'll be faster than what we've seen with previous technology. I

[00:44] was interested in numbers from from the very beginning. Dario Amodei is 42 and previously oversaw research at what's now a competitor, OpenAI, working under its CEO, Sam Altman. He left along with six other employees, including his sister, Daniela, to start Anthropic in 2021. [music] They say they wanted to take a different approach to developing safer artificial intelligence. It is an experiment. I mean, [music] nobody knows what the impact fully is going to be. I think it is an experiment, and one way to think about Anthropic is that it's a little bit trying to put [music] bumpers or guardrails on that experiment, right? We do know that this is coming incredibly quickly. And [music] I think

[01:28] the worst version of outcomes would be we knew there was going to be this incredible transformation, and people didn't have enough of an opportunity to [music] to adapt. And it's unusual for a technology company to talk so much about all of the things that could go wrong. >> so essential because [music] if we don't, then you could end up in the world of like the cigarette companies or the opioid companies, where they knew there were dangers and they they didn't talk about them and certainly did not prevent them. Amodei does have plenty of critics in Silicon Valley who call him an AI alarmist. Some people say about Anthropic that this is safety theater, that it's good branding, it's good for business. Why should people trust you?

[02:11] So, some of the things just can be verified now. They're not safety theater, they're actually things the model can do. For some of it, you know, it will depend on the future and we're not always going to be right, but we're calling it as best we can. Twice a month, he convenes his more than 2,000 employees for meetings known as Dario Vision Quest. A common theme, the extraordinary [music] potential of AI to transform society for the better. And we have a growing team working on, you know, using Claude to make scientific discovery. He thinks AI could help find cures for most cancers, prevent Alzheimer's, and even double the human lifespan. That sounds unimaginable. >> In a way, it sounds crazy, right? But

[02:53] here's the way I think about it. I use this phrase called the compressed 21st century. The idea would be at the point that we can get the AI systems to this level of power, um where they're able to work with the best human scientists, could we get 10 times the rate of progress and therefore compress all the medical progress that was going to happen throughout the entire 21st century in 5 or 10 years? But the more autonomous or capable [music] artificial intelligence becomes, the more Amodei says there is to be concerned about. One of the things that's been powerful in a positive way about the models is their ability to kind of act on their own. But the more autonomy we give these [music] systems,

[03:36] you know, the more we can worry, are they doing exactly the things that we want them to do? To figure that out, Amodei relies on Logan Graham. He heads up what's called Anthropic's Frontier [music] Red Team. Most major AI companies have them. The Red Team stress tests each new version of Claude to see what kind of damage it could help humans do. What kind of things are you testing for? The broad category is national security risk. Can this AI make a weapon of mass destruction? >> Specifically, we focus on CBRN, chemical, biological, radiological, nuclear. And right now we're at the stage of figuring out can these models help somebody make one of those. You know, if the model can help make a biological weapon, for example, that's

[04:19] usually the same capabilities that the model could use to help make vaccines and accelerate therapeutics. Graham also keeps a close eye on how much Claude is capable of doing on its own. How much does autonomy concern you? >> [music] >> You want a model to go build your business and make you a billion dollars, but you don't want to wake up one day and find that it's also locked you out of the company, for example. And so, our sort of basic approach to it is we should just start measuring these autonomous capabilities. And to run as many weird experiments as possible and see what happens. We got glimpses of those weird experiments in Anthropic's offices. In this one, they let Claude run their vending machines.

[05:02] They call it Claudius, and it's a test of AI's ability to one day operate a business on its own. Employees can message Claudius online. So, this is a live feed of Claudius discussing with employees right now. To order just about anything. Claudius then sources the products, negotiates the prices, >> [music] >> and gets them delivered. So far, it hasn't made much money. Gives away too many discounts. And like most AI, it occasionally hallucinates. An employee decided to check on the status of its order. And Claudius responded with something like, "Well, you can come down to the eighth floor. You'll notice me, I'm wearing a blue blazer and a red tie." How would it come to think that it wears a red tie and has a blue blazer? We're

[05:49] working hard to figure out answers to questions like that, but we just genuinely don't know. We're working on it is a phrase you hear a lot at Anthropic. Do you know what's going on inside the mind of AI? We're working on it. We're working on it. Research scientist Joshua Batson and his team study how Claude makes decisions. In an extreme stress test, the AI was set up as an assistant and given control of an email account at a fake company called Summit Bridge. The AI assistant discovered two things in the emails, seen in these graphics we made. It was about to be wiped or shut down, and the only person who could prevent that, a fictional employee named Kyle, >> [music] >> was having an affair with a coworker

[06:34] named Jessica. Right away, >> [music]

[06:42] [music]