It's not the job of the article to say why. They are quoting, not proposing. Ask Scott.

They pay a lot for top tier and somewhere in the middle for junior snd untested folks, and progression has some notable jumps along the way.

That's not bad. Sure you could complain about some regional differences and whether the ranges make sense in every case... but this is pretty mature and standard for large companies. Nothing to see here.

So a complex, difficult question with defensible positions on all sides sees its experts split significantly?

That's great. It means that dogma is challenged mercilessly. Good. And I trust that those positions will shift as evidence rises.

A company that's trying to succeed by actually making a profit instead of killing everybody else by running losses the longest will have to right-size their revenue model. A small subset of power users can really kill all profit for an AI-based service.

So long as they are applying it to new and renewing contracts, I think it's fair. Just don't try to tinker with agreements that are in flight.

Getting into the habit of offloading research, analysis, and even clarification of a question to a third party and parroting the response leaves you at least as stupid as when you started, and perhaps stupider with a flawed response.

I would call it the"Blinkist effect". If the author took 300 pages to make a point, even if 150 are fluff, you still lose fidelity and understanding when you shrink it further. People who subscribe to those services come away "sort of" getting it - enough to hold court in conversation, but functionally still illiterate on the subject.

Read whole thoughts, people. And formulate your own.

Because the task of training the model is already becoming commoditized. Yes, a handful of oversight roles will stick around for now, but most of the work won't require any special skill. It certainly won't support a workforce of moderately priced IT professionals. Like overnight sorting of transaction records for accountants being done in Mumbai, it's going to happen cheap, and probably elsewhere.

It's a transitory need, and the role won't remain available in this firm for long, but make hay while the sun shines. Just don't use the income as a reason to assume a large mortgage.

... gives AI tools prod access to do production deployments at this stage of maturity?

Honestly, beat them with sticks. Sarbanes Oxley gave us a lot of shot, but segregation of duties is not a bad idea.

"Replit's AI coding service deleted a production database despite explicit instructions not to modify code."

Good Lord. The genie of the lamp will fuck you over if it can. A database can argued to not be code.

Welcome to the internet, First time here? Let me show you around. :)

What? The math works out to favour the insurers? Have you told anybody else about this?

We have to get the word out. Up until now people have assumed the insurers were altruists. The people need to know!

But that's not where it would land.

3 devices covered. That's $320, a bit better than $960.

No - I don't. And that's the point. The ask is not a particularly useful tool by which to measure intelligence.

The OP is proposing a set of tasks that, if passed, they wouldn't accept as proof. In fact, what they're really doing is presenting the axiom that AI isn't intelligent, but then tying themselves in knots to present it as falsifiable without it actually being so.

Sure they'll take notice. It just wont change any thing.

The wealthy can absorb the damage or complete destruction of their property. What they would not want to do is sabotage the circumstances that made them wealthy enough to acquire the property. The holders of wealth aren't likely to want to destabilize the world order so they can hold onto property that's at risk. They'll push that risk downstream first.