lexandstuff

Seems like a pretty big step towards understanding the language no?

I remember Experts Exchange somehow managed to have every question one could ever ask, but no actual answers - even with a paid account.

I was glad when S.O. put an end to them.

I prepared my taxes this year using Astra + Codex. It did a thorough job on a complicated return, and I sent it to my accountant to fill out the tax form, with everything laid out in a spreadsheet. They sent it back for final review, and I had Codex check it; it was riddled with errors. The refund was nearly $2k lower than it should have been, and they put a bunch of stuff into the wrong fields. Had to go back and forth multiple times to get it 100% correct. Made me wonder how many previous years' returns were wrong.

I feel like we're going to get to a point where it's mandatory to use AI for certain tasks - humans just can't be trusted to get things right that require a level of precision.

In a lot of ways, robotics - navigating and operating in the physical world - seems to be a very verifiable problem. It's fairly easy to verify that a robot moved from A to B, or that it built a structure that completely aligns with the plan, for example.

The main issue is cost and speed to verify, but simulations and world models will help there. I think we'll start seeing rapid progress pretty soon.

I worked in a mattress factory for a while, doing the same repetitive task day in and out.

I remember feeling like the days passed painfully slowly, but the years flew by.

To the contrary: Twitter has the lowest content engagement rates out of any of the networks.

Yep.

LinkedIn is pretty much the only social network out there where the average person has a shot at a little reach for their posts.

Even if it’s just past and present colleagues and various acquaintances, I actually have an audience of people who are sometimes interested in what I have to say. I can't say that for any other social network.

Yes, MCP is an API.

Previously, it was an API with a weird stateful standard; now it is essentially a bog-standard stateless JSON-RPC API with conventions for tool discovery, long-running tasks and multi-round-trip requests

If you wrote your own JSON-RPC API for an LLM, you would probably eventually end up with something that looks very much like the standard they're describing.

It should be SotA.

Ilya Sutskever said in 2017, whilst building OpenAI:

> Within the next three years, robotics should be completely solved, AI should solve a long-standing unproven theorem, programming competitions should be won consistently by AIs, and there should be convincing chatbots.

There was a viral YouTube video from 2014 called Humans Need Not Apply [1], which made a very strong argument that mass unemployment was right around the corner.

I actually think people from the past would be surprised at how slow AI has progressed in the end. Although the domains where it's turned out to be most effective - code, images, videos, music, etc. - would probably surprise them.

[1] https://www.youtube.com/watch?v=7Pq-S557XQU

Thank you!

I take your point - no doubt I approached this in a very naive way.

That said, we did collaborate on it - at the very least I needed to learn his address before sending.

Neither of us have ever sent or received a package in Uganda. It was a learning experience for both of us.

~$200 doesn't go as far as you'd expect for good used laptop, even in Uganda. We did look into our options.

However, there's definitely a sunk cost aspect to the operation. After the first failure to send it through Australia Post, I became determined that Django was going to have that MacBook.

I have a feeling that a big risk of using AI all the time is that our own neurological capacity starts to dwindle.

Just as many people leading sedentary lifestyles have to make a deliberate effort to exercise, because inactivity is really bad for our bodies, I think we're going to realise that a similar process is necessary for our minds.

You really want to be spending a bit of time every day operating at your cognitive limits - trying to fully engage your System 2 - if you want to avoid brain atrophy. Coding used to kind of give you this exercise for free, but you can go really far with just your System 1 nowadays - literally get things done while scrolling Reddit.

I'm trying to allocate 30-60 minutes a day to doing something difficult, like writing code by hand for an unfamiliar problem or reading and summarising difficult papers without AI.