Not kidding, I bet llm’s are excellent at triaging these reports. Humans, in a corporate setting, are apparently not.

It does if you’re a clumsy operator and those are not rare.

Yes, but the machine itself is deterministic and logically sound.

>Yes, but the machine itself is deterministic and logically sound.

Because arithmetic itself, by definition, is.

Human language is not. Which is why being able to talk to our computers in natural language (and have them understand us and talk back) now is nothing short of science fiction come true.

Even worse is if it's in the other room and your fingers can't reach the keys. It delivers no answers at all!

My point is, needing to use something with care doesn't prevent it becoming from wildly successful. LLM's are wrong way more often but are also more versatile than a calculator.

> LLM's are wrong way more often but are also more versatile than a calculator.

LLMs are wrong infinitely more than calculators, because calculators are never wrong (unless they're broken).

If you input "1 + 3" into your calculator and get "4", but you actually wanted to know the answer to "1 + 2", the calculator wasn't "wrong". It gave you the answer to the question you asked.

Now you might say "but that's what's happening with LLMs too! It gave you the wrong answer because you didn't ask the question right!" But an LLM isn't an all-seeing oracle. It can only interpolate between points in its training data. And if the correct answer isn't in its training data, then no amount of "using it with care" will produce the correct answer.

What about a strict subset of C#. The use case for F# seems to be shrinking because MS is putting all its energy in language.

This isn't true, because more is not always better.

C# has lots of anti-features that F# does not have.

This seems like an actually useful computation to do, unlike earlier results. Is that a reasonable reading of this article?

No it’s still completely useless for the real world. Also not actually verifiable

Classic quantum

Now verifiably useless in real life

An AI powered meal planner that helps you create recipes, plan your weeks and manage your groceries:

https://github.com/bobjansen/mealmcp

There is a website too so you don’t actually need to use MCP:

https://meals.bobjansen.net/

I understand this as follows:

Legal: doesn’t break the rules of chess. For example: no pawns on the eighth rank or in check when it’s not their turn.

Reachable: there is a series of moves that lead to this position from the standard starting position.

This “feature” amazes me. It is badly done and a bad idea. I haver never watched a dubbed video so why show me a translated title? It’s also surprising, Google has plenty ESL employees on staff.

There has to be some KPI tied to how often the AI model is used in production for providing translations on YouTube etc. Someone's promotion hangs on the translation feature being used as often as possible on YouTube.

The point is that bit of information doesn’t tell you anything about yourself and your own skills and intrinsic value. You could be there best hire to date but someone even better than you shows up, or the bosses nephew or funding went away, etc. All events outside of your control.

Really, what are you making that a 5% increase in productivity doesn’t justify a Copilot subscription?

That's not a rigorously measured number.

The 5% is an increase in straight-ahead code speed. I spend a small fraction of my time typing code. Smaller than I'd like.

And it very well might be an economically rational subscription. For me personally, I'm subscription averse based on the overhead of remembering that I have a subscription and managing it.

There is a Wikipedia page:

https://en.wikipedia.org/wiki/Satellite_collision