sohois
7d ago
I agree on the Shield. I think it came out just a bit too early to really benefit from being 'prestige TV'. Breaking Bad's popularity exploded in later seasons thanks to the growth of the internet and social media, but more than that it had the room to focus on Walt. The Shield was too long, bogged down with a lot of aimless plotlines; it felt like the production was still caught between making a serious character study, and a procedural cop show at times.
sohois
7d ago
Claude managed to peg me again
Finally AI is delivering what the people really want
sohois
8d ago
An LLM, strictly speaking, can't do that
That depends on how specific we want to be with "LLM". I would be surprised if ChatGPT did not have superior chess performance to someone directly calling GPT 5.4 and giving it no scaffolding beyond "We're playing chess", but few people would argue that using ChatGPT is giving an LLM extra capabilities. How much of a harness is appropriate?
If you give Claude or Codex access to a notepad skill to record any chess moves, their performances would improve. If you gave them a simple chess application with a virtual board to record moves, it would probably get even better. Would you still say that an LLM can't "strictly" augment itself in this situation?
As you yourself have pointed out, the LLM you played your simple game with independently proposed maintaining a log of game state within text, so it would be fair to say that LLMs can recognize and attempt to remedy weaknesses without human intervention, just as a normal person would do
sohois
8d ago
The chess argument is not at all a good analogy, because like a huge bunch of AI criticism, it vastly overstates normal human capacity.
What's the biggest reason an amateur teenager doesn't make impossible moves? It's because they have a chess board right in front of them. It's extremely easy to track the state of the game and position of pieces when you have the laws of physics doing it for you. How many amateurs do you think could perfectly recreate a given game state if someone came along and threw the board over?
LLMs play chess entirely through text. It's the equivalent of asking a person to play a game of correspondance chess, buth they can't recreate the game physically, they can't have any drawings of the game, all they can do is have a record of moves already made. Outside of literal chess masters, how many humans would get through such a game without making a mistake?
sohois
11d ago
AI agents are still in that awkward spot: "I spent 3 hours automating a task that would have taken me 5 minutes manually". Except previously only those skilled with computers could try this on a small range of tasks, and now anyone can do it for pretty much anything.
You absolutely could spend a week setting up Claude or Codex or whatever, putting in all the connectors and skills you need, and running a bunch of trial runs to fully automate a big chunk of your job. You'll still end up bottlenecked by stuff. Depends how much you actually enjoy working I suppose
sohois
11d ago
2 and 3: I expect the Claude Chrome add-on could handle those, depending on who each website is set-up. Two issues though: first, Claude is actually pretty slow going through Chrome, and no guarantee it wouldn't be faster to just do it yourself. Second, you're opening yourself up for all kinds of security risks at the moment.
1: As a below post says, agents can now use computers directly, but it's not great yet. Not knowing much about the situation, I expect an MCP would be enough, but it heavily depends on the software.
Tbh, the easiest solution would be to see if any of these have some kind of "Export to csv" option, then just chuck those files at Claude/ChatGPT/Whatever and tell it what you need
sohois
15d ago
I saw this Code setup in one Zvi's roundups. Might be a bit OTT for your uses though:
sohois
15d ago
Despite the name, it isn't really coding specific. I presume "Claude cowork" is largely the same as Claude code but with a name that doesn't scare the hos.
From what I can recall, Cowork was announced back when Claude Code was still only available through the terminal, and Cowork was the easy ho-accessible version. And then they make Claude Code available through the Claude app anyway, so not really clear what differences there are anymore
sohois
28d ago
Regardless, a few developments have happened recently that have motivated me enough to actually make a top-level post about this. The first being my (employer-mandated) use of Claude to generate code. "You're not using the latest model, just one more model and we'll reach AGI"-bros officially in shambles after this one
This was with the latest version of Claude Sonnet
>I totally used the latest version and it still sucked!!
>Look inside
>Not the latest version
I'm not even gonna claim Opus would make a huge difference because the differences are quite small at this point, but fuck me you would think you might have a little humility when making such an emphatic claim, just to contradict yourself within a paragraph.
Just ignore the mails. You can go online and state that you don't need a licence, but it's easier to just do nothing, and ignore any attempts at contacting you. The enforcement group has zero power and it's basically impossible for them to ever demonstrate you have watched the BBC unless you literally let them in your house while you're watching it.
More options
Context Copy link