aaronrobinson

Isn’t a didge just a big straw anyway?

But somehow the conversation is still enthralling. Just seeing the first few words of each response gives you a feel for the level they’re on.

“ Let the frontier labs win by being better; don’t let them define safety or security, or pull up the ladder of humanity’s collective knowledge”

Love this.

I didn’t suggest it was a problem?

For those downvoting this are you doing so because I tried to pass it through an LLM, I commented that it was hostile to AI, because it's being hostile or something else?

I think it's been set up to server some other content if LLMs access the page as they both complained about seeing more general ARIA stuff. But copying and pasting the page into Claude was what detected the instruction to AI to not allow summaries but Claude overruled it and said my request was more important!

I tried to pass the page through Claude/ChatGPT to ask questions about the article. The AI hostility is real! This is the first time I’ve seen an instruction to AI in a web page that asks AI not to summarise.

This is amazing

Nice read. Very interesting.

Begs the question why they didn’t present that info to Anthropic directly and if they did why they didn’t act

What made you choose Torino?

Attenborough is everything the US doesn’t have.

They look guilty in that picture

The title makes it sound like they’re removing the remains of lost Lords gathering dust on the seats although that’s probably not too far from the truth.

You sound like an absolute nightmare to work for.

Me too. I’m loving it.

Why wouldn’t you ask AI to explain the architecture and code? It’s much better and efficient than any human.

I was thinking about cancelling my OpenAI subscription. This will do it.

Or using ads

We have no cure now but that may change and depend on early detection just as taking meds now can slow the onset. I don’t want to know makes no sense to me. You would plan your whole life differently and it would actually be quite liberating once you’d come to terms with it.

5) US Gov uses a different law 6) Go to step 1

It’s not to feel important, it’s to make others feel they’re important. This is the definition of corporate.

Rooibus. It’s not decaf, it just has no caffeine. I drink it all the time since giving up coffee.

You need a mix of price, availability and trust. I just head for the nearest Sainsbury’s. They have. Reputation to protect, I can do my shopping at the same time and their prices are competitive. I just wouldn’t go to the cheapest place unless it was a hire car.

What a dude.

I read that people just allow Claude Code free rein but after using it for a few months and seeing what it does I wonder how much of that is in front of users. CC is incredible as much as it is frustrating and a lot of what it churns out is utter rubbish.

I also keep seeing that writing more detailed specs is the answer and retorts from those saying we’re back to waterfall.

That isn’t true. I think more of the iteration has moved to the spec. Writing the code is so quick now so can make spec changes you wouldn’t dare before.

You also need gates like tests and you need very regular commits.

I’m gradually moving towards more detailed specs in the form of use cases and scenarios along with solid tests and a constantly tuned agent file + guidelines.

Through this I’m slowly moving back to letting Claude lose on implementation knowing I can do scan of the git diffs versus dealing with a thousand ask before edits and slowing things down.

When this works you start to see the magic.

It can definitely feel like that right now but I think a big part of that is us learning to harness it. That’s why resources like this are so valuable. There’s always going to be pain at the start.

What a strange response.

You may have a point but you lost me citing Vance and Musk.

This is the same guy that created the ZX81 (that I learned to code on) and the ZX Spectrum. He changed my life.