aliston

I'm having trouble just keeping track of all these different types of models.

Is "Gemini 3 Deep Think" even technically a model? From what I've gathered, it is built on top of Gemini 3 Pro, and appears to be adding specific thinking capabilities, more akin to adding subagents than a truly new foundational model like Opus 4.6.

Also, I don't understand the comments about Google being behind in agentic workflows. I know that the typical use of, say, Claude Code feels agentic, but also a lot of folks are using separate agent harnesses like OpenClaw anyway. You could just as easily plug Gemini 3 Pro into OpenClaw as you can Opus, right?

Can someone help me understand these distinctions? Very confused, especially regarding the agent terminology. Much appreciated!

Wait, are you saying the Obama O’s story - that the 500 boxes sold out, saving AirBnb - is not totally true?

You can find a medical professional to basically claim anything these days. I could go into specifics, but there's a whole industry of ethically questionable doctors that can help you take advantage of well-intentioned accommodation policies with a subjective diagnosis. While I agree there are cases of serious stress disorders, there are also a bunch of people claiming a disorder for personal benefit.

There's also a huge difference between someone who is working as an ML researcher versus someone working on AI infra or AI-based feature development.

Wow, I hadn't thought about it through that lens exactly, but you are 100% right. Virtually everyone in large organizations (engineers, PMs, designers) are incentivized to launch things that align with their end-of-year performance goals. Those goals are sometimes at odds with the best overall customer outcome.

I hadn't thought about figma specifically as being a symptom. I'm sure some organizations use it effectively, but I can see how it might spiral out of control to result in an "I'm helping too" sort of ethos, with a poor net outcome.

What's so bad about the Spotify app? For the most part, it looks like a bunch of lists (not so much spreadsheets) to me...

As someone that is only tangentially in the design space, why do you think collaborative design works so poorly? I have noticed that, in engineering reviews, complex backend designs get scrutiny, but rarely "I feel..", "I prefer..." types of comments, whereas frontend teams get all types of those comments. Is it a matter of too many cooks in the kitchen or something else?

While Faraday discovered induction, wasn't it Maxwell that unified electricity and magnetism? Given what answer.ai is attempting to do, Edison seems like a great example since he was both a brilliant inventor and an absolutely shrewd businessman.

I am excited for more research in this area, since there is currently a huge gap between foundational model research and practical applications of AI.

I don't know if it was led by Sam, and don't dispute that it may have been "hacky," but there is no denying it was a visionary project.

Yes, other companies had similar models. I know Google, in particular, already had similar LLMs, but explicitly chose not to incorporate them into its products. Sam / OpenAI had the gumption to take the state of the art and package it in a way that it could be interacted with by the masses.

In fact, thinking about it more, the parallels with Steve Jobs are uncanny. Google is Xerox. ChatGPT is the graphical OS. Sam is...

This is actually a semi-plausible angle. Given Sam's personality, I could see a scenario where there was disagreement about whether something in particular would be announced at demo day. He may have told some people he would keep in under wraps, but ended up going forward with it anyway.

I don't understand how that escalates to the point that he gets fired over it, though, unless there was something deeper implied by what was announced at demo day.

Edit: Theres a rumor floating around that "it" was the GPT store and revenue sharing. If that's the case, that's not even remotely a safety issue. It's just a disagreement about monetization, like how Larry and Sergey didn't want to put ads on Google.

ChatGPT blew the doors open on the AI arms race. Without Sam leading the charge, we wouldn't have an AI boom. We wouldn't have Google scrambling to launch catch up features. We wouldn't have startups raising 100s of millions, people talking about a new industrial revolution, llama (2), all the models on hugging face or any of the other crazy stuff that has come about in the past year.

Was the original launch of ChatGPT "safe?" Of course not, but it moved the industry forward immensely.

Swisher's follow up is even more eyebrow raising: "The developer day and how the store was introduced was in inflection moment of Altman pushing too far, too fast. My bet: He’ll have a new company up by Monday."

What exactly from the demo day was "pushing too far?" We got a dall-e api, a larger context window and some cool stuff to fine tune GPT. I don't really see anything there that is too crazy... I also don't get the sense that Sam was cavalier about AI safety. That's why I am so surprised that the apparent reason for his ousting appears to be a boring, old, political turf war.

My sense is that there is either more to the story, or Sam is absolutely about to have his Steve Jobs moment. He's also likely got a large percentage of the OpenAI researcher's on his side.

If it was really just about seeing eye to eye, why would the press release say anything about Sam being "consistently candid in his communications?" That seems pretty unnecessary if it were fundamentally a philosophical disagreement. Why not instead say something about differences in forward looking vision?

I've been out of the web-dev game for a while, and have been pretty blown away by how much has changed. Suddenly, it seems like everything is a 1-page application, SSR, and doing a lot more on the client than I was used to back when jQuery was the standard. Next.js seems like a pretty beginner-friendly way to get a lot of the "new stuff" without having to learn the multiple technologies in a MERN stack.

On the other hand, it seems like a lot of folks (particularly in this thread), think of Next.js as a well-intentioned, but unstable shiny toy that more or less breaks when you try to do more complicated stuff. Is that accurate? If so, what is the standard stack that most companies are using these days to build high quality web apps?

Ironically, I was just using ChatGPT to answer another seemingly simple question:

What is an example of an item that would show up in EBIT but not operating income?

> In standard financial reporting and accounting terminology, "Operating Income" and "EBIT" (Earnings Before Interest and Taxes) are typically considered to be the same thing. They both represent a measure of a company's profitability from its core operating activities, and they exclude non-operating income and expenses...

It's an answer, stated as fact, that is also totally wrong. When I went to Google, I found the correct answer:

"The key difference between EBIT and operating income is that operating income does not include non-operating income, non-operating expenses, or other income."

After some pressing of ChatGPT, I ultimately got it to spit out the correct answer:

"I apologize for any confusion in my previous responses, and you are correct. In standard financial reporting, EBIT does include non-operating income, whereas operating income does not."

ChatGPT is great in some cases, and in other cases sends you down random tangents, references things that either don't exist or are misleading, or just flat out lies. That is a REAL time waster, because you might end up proceeding based on the incorrect explanation provided by ChatGPT, only to find out that you have to re-learn a concept correctly later on.

Is it true that the safeguards are considered part of the model? I had assumed that the "safeguards" that limit certain types of responses in ChatGPT were separate from the actual language model.