jjfoooo4

I expected Kimmel to have somehow criticized Kirk, a dubious enough reason to pull the show. But this isn’t even that. Comments quoted in stories assert that the shooter was MAGA - maybe that’s somewhat controversial, but it’s ludicrous to suggest it’s offensive.

That paired with comments criticizing the Dear Leader were enough. This is a new low in corporate cowardice toward Trump bullying.

Why do you think OpenAI wanted to get rid of GPT-4 etc so aggressively?

I suppose there's a distinction between new less capable models, I can see why those would be more efficient. But maybe the older frontier models are less efficient to serve?

I think what we'll eventually see is frontier models getting priced dramatically more expensive (or rate limited), and more people getting pickier about what they send to frontier models vs cheaper, less powerful ones. This is already happening to some extent, with Opus being opt-in and much more restricted than Sonnet within Claude Code.

An unknown to me: are the less powerful models cheaper to serve, proportional to how much less capable they are than frontier models? One possible explanation for why e.g. OpenAI was eager to retire GPT 4 is that those older models are still money losers.

The article claims otherwise:

> In fact, even if you remove the cost of training models from OpenAI's 2024 revenues (provided by The Information), OpenAI would still have lost $2.2 billion fucking dollars.

People using their free offering for personal use is a critical part of Notion's sales pipeline. It's a hell of a lot easier to sell a piece of software to a company when all it's internal users have already trained themselves how to use it in their personal time. Those users are probably the ones generating the enterprise sales lead in the first place

See also: Google Docs

There's certainly a symbiosis blog publishers and small startups wanting to be perceived as influential, and big companies releasing models and wanting favorable coverage.

I heavily discount same day commentary, there's a quid pro quo on early access vs favorable reviews (and yes, folks publishing early commentary aren't explicitly agreeing to write favorable things, but there's obvious bias baked in).

I don't think it's all particularly concerning, you can discount reviews that are coming out so quickly that's it's unlikely the reviewer has really used it very much.

I mean, not getting confused / forgetful is a pretty big one!

I think one thing it does is help you get rid of the UX where you have to manage a bunch of distinct chats. I think that pattern is not long for this world - current models are perfectly capable of realizing when the subject of a conversation has changed

Yeah to some degree that's already happened. Anecdotally I hear giving your whole iMessage history to Gemini results in pretty reasonable results, in terms of the AI understanding who the people in your life are (whether doing so is an overall good idea or not).

I think there is some degree of curation that remains necessary though, even if context windows are very large I think you will get poor results if you spew a bunch of junk into context. I think this curation is basically what people are referring to when they talk about Context Engineering.

I've got no evidence but vibes, but in the long run I think it's still going to be worth implementing curation / more deliberate recall. Partially because I think we'll ultimately land on on-device LLM's being the norm - I think that's going to have a major speed / privacy advantage. If I can make an application work smoothly with a smaller, on device model, that's going to be pretty compelling vs a large context window frontier model.

Of course, even in that scenario, maybe we get an on device model that has a big enough context window for none of this to matter!

There are many folks working on this, I think at the end of the day the long term memory is an application level concern. The definition of what information to capture is largely dependent on use case.

Shameless plug for my project, which focuses on reminders and personal memory: elroy.bot

But other projects include Letta, mem0, and Zep

If it's of interest, I expanded on this in a blog post: https://elroy.bot/blog/2025/07/29/ai-is-a-floor-raiser-not-a...

It's a floor raiser, not a ceiling raiser. It helps you get up to speed on general conventions and consensus on a topic, less so on going deep on controversial or highly specialized topics

There was also a recent article on the topic in the WSJ, with an interview from the DeBeers CEO:

https://www.wsj.com/business/retail/de-beers-diamonds-price-...

In it he calls lab grown diamonds a "huge con", though doesn't really explain why...

Is that not what "Chill Mode" is? I've activated that on my car to relieve some of the jerkiness and it seems to have worked well

Wouldn't that behavior already be reflected in these numbers?

I also just activate the desired virtual env and start VS Code from the command line within it

I've gotten a lot of utility out of LLM Cmd (https://github.com/simonw/llm-cmd), but one thing I didn't like was that generated commands don't show up in my shell history.

I wrote a small shell function that loads the output commands as user input rather than executing them directly. This way, you can search through generated commands later!

X didn't "invest in AI", it was rolled into a buzzy AI company. Before that the holders of it's debt could not find buyers (aka buyers willing to bet against X bankruptcy)

I think Elon truly believed in the subscription model, which would free him from advertiser content influence. That and being terminally addicted to the platform himself, and being an impulsive gambler. I really don't think we've gotten where we are due to any (successful) master plan

I don’t think the comments here are complaining, they are pointing out that what’s being claimed as being new is not actually new

It's very hard for me to understand what MCP solves aside from providing a quick and dirty way to prototype something on my laptop.

If I'm building a local program, I am going to want tighter control over the toolsets my LLM calls have access to.

E.g. an MCP server for Google Calendar. MCP is not saving me significant time - I can access the same API's the MCP can. I probably need to carefully instruct the LLM on when and how to use the Google Calendar calls, and I don't want to delegate that to a third party.

I also do not want to spin up a bunch of arbitrary processes in whatever runtime environment the MCP is written in. If I'm writing in Python, why do I want my users to have to set up a typescript runtime? God help me if there's a security issue in the MCP wrapper for language_foo.

On the server, things get even more difficult to justify. We have a great tool for having one machine call a process hosted on another machine without knowing it's implementation details: the RPC. MCP just adds a bunch of opinionated middleware (and security holes)

What is the point of a MCP server? If you want to make an RPC from an agent, why not... just use an RPC?

Anthropic's marketing has been successful at position them as the responsible alternative to OpenAI, but what concretely do they actually do differently than any other model provider?

It feels very akin to the Uber vs Lyft situation, two companies with very different perceptions pursuing identical business models

The AI executives predicting AI doomsday trend has been pretty tiresome, and I'm glad it's getting push back. It's impossible to take it seriously given an Anthropic's CEO's incentives: to thrill investors and to shape regulation of competitors.

The biggest long term competitor to Anthropic isn't OpenAI, or Google... it's open source. That's the real target of Amodei's call for regulation.

I like the bit of humor at the start of the video, but the bit goes on for a bit too long

I think more specifically, if you don't have strong guidance to the LLM on how a project is organized and where things should go, things go pretty far off the rails.

I tried to have an LLM fully write a Python library for me. I wrote an extensive, detailed requirements doc, and it looked close enough that I accepted the code. But as I read through it more closely, I realized it was duplicating code in confusing ways, and overall it took longer than just getting the bones down myself, first.

Some coding agents are now more actively indexing code, I think that should help with this problem

> How is someone just coming out of school going to get the encouragement and space to independently develop the experience they need to break out of the "vibe coding" phase?

LLM's are so-so coders but incredible teachers. Today's students get the benefit of asking copying and pasting a piece of code into an LLM and asking, "How does this work?"

There's a lot of young people that will use LLM's to be lazy. There's also a lot that will use them to feed their intellectual curiosity.

It was about very stupidly thinking that selling subscriptions to Twitter would be more profitable than selling ads

MCP is described as a means to make the web open, but it’s actually it’s a means to make demos of neat things you could do if the web were actually open.

I’ve been seeing MCP compared to extensions in web browsers. Which I find telling, since I wouldn’t exactly say web extensions have been a great success - it’s a pretty niche dev market, and the security posture remains pretty anxiety inducing

I think updating the former to the latter when you are actually debugging something isn’t that big of a deal.

But with actually checked in code, the tradeoff in readability is pretty substantial