>While this underlying RT, found in a jumbo phage, had been identified in previous studies, Claude appears to be the first to notice the system’s defining features—an associated array of non-coding DNA sequences and an additional accessory protein of unknown function.

So they investigated an already known thing. Not exactly "discovering a new system"... Anyone with money to throw at this already-known thing would have gotten those results I assume.

I feel this is overly pessimistic. Perhaps see it this way then; it is now easy and cheap to throw money at a Thing.

Society is bottlenecked by the limited amount of experts it can muster. That is increasingly less the case.

Money and people and realizing in advance that this particular thing is worth concentrating upon, out of a thousand or maybe a million other opportunities.

Even the people parameter is a serious limitation, in all sorts of domains. An example: we have a huge stash of ancient cuneiform tablets from the Middle East, but most have not been read yet because there are very few people who are able to read them.

Isn't it the whole point?

Throwing money at a problem was expensive, it's a lot less expensive now

If we have AGI, it can just solve all the memory issues in the existing codebase instead of rewriting it all.

But then we'd still be left with a C or C++ codebase.

Given the hypothetical presupposes AGI, so what? We wouldn't be the ones reading or maintaining it.

We have no idea how many of these engineers work on the mobile app. This comment is somehow both AI slop and human slop.

Google is guilty of this across their whole range of product. We hear a lot about Youtube, but it's the same for Ads and Android. I had my developer account banned for a copyright issue, and to this day I don't know what the issue actually was.

Google has always been like this. I used to have a Youtube account with timelapse movies from my holidays. Nothing special, just stuff to show the family, but it was a lot of work to edit and upload them all.

One day I got a message that my account was closed because of a violation of the TOS. No explanation what it was that I violated. No chance for appeal. No chance to communicate with a human about it. It was explicitly stated that it was impossible to reinstate the account. To this day I have no clue what it was in the 20 or so timelapse movies taken from a small sailing boat that offended Google.

It was a good lesson to never trust Google.

“If [virtualization is] not built into a broader AI operation stack, the value is even lower again,” Pickering said.

For the love of god keep Gen AI off of infrastructure...

>The harder question, in Chollet’s framing, is: how efficiently can a system learn to do something genuinely new?

The more diverse stuff it knows, the easier it will be to learn something new.

While humans can do this, it has historically been difficult for machine learning models to manage it.

It is unclear if Transformers are "it" or not, because while they are much more general, they are also very spikey intelligences despite having read almost everything the trainers can get their hands on.

The people that go through the trouble of installing apks from random websites to not pay $1 are in the minority. What's more, most of them wouldn't pay that dollar if they couldn't find the cracked apk. So the only thing you're realistically doing is cutting yourself off of a huge market. Convenience sells, the Play Store does that

$1 - most probably. What about $20?

Unless your target market is a small niche of hackers, then yeah the vast majority would pay $20.

Just look at how many pay for music and video streaming services, when it's trivial to download anything for free from illicit sources.

You already have an agent freely doing stuff on your machine. Subagents prompts are a weird place to draw a line. It's not like you're reading everything the agent is doing in any case, let's not kid ourselves.

When things go wrong I very much read the session traces to figure out what in my prompt wasn't good/explicit enough, then retry to evaluate if it would have helped.

I was about to do the same with Sol + Ultra, but then discovered this encryption issue that prevents me from doing the same for sub-agents.

> It's not like you're reading everything the agent is doing in any case

Personally I do, these tools aren't mature enough to be used without supervision

> You already have an agent freely doing stuff on your machine

No. Agents run in VMs. Assume anywhere you’re running an agent will be compromised, because eventually, it will be.

The only reason most people haven’t is luck, they didn’t happen to install Axios or Tanstack at a certain time.

It's useful to show the actual team that it's possible. From there, they can make the decision of whether to go the bun route with more information.

I'm not sure I see the value in "showing the team it's possible". I would presume the team are intelligent enough to be well aware that it's possible, given tradeoffs. And as it's clear those tradeoffs have been "traded" in this case, it doesn't seem like having the knowledge confirmed is particularly surprising.

This is true but these projects are rarely presented or interpreted as a proof of concept.

The bun route == spending 100k for an LLM rewrite as an advertisement for the LLM company you sold your soul to.

> I’m happy to explain why an essential app is worth your money

This is a podcast app. It's in no way essential, how did you use to explain this to your customers ?