Hypes

This is a new paradigm for interacting with Claude that is significantly more "inline" with all the other human activity org-wide. Once you do all of the under the hood engineering work to make this "just work" (e.g. across tools, integrations, compute environments, memory, security, etc.), Claude basically joins the team in a seamless way - you can talk to it as you would talk to a person and it can help with a very large variety of workloads. Imo this is the 3rd major redesign of LLM UIUX. The first paradigm was that the LLM is a website you go to, the second was that it is an app you download to your computer. This third one is that it is a self-contained, persistent, asynchronous entity with org-wide tools and context, working alongside teams of humans. It really takes a while to wrap your head around it, but it works and it is awesome.

@karpathy

rumors of new models being delayed

@GaryMarcus

All Mythos-level models are likely to invite similar risks. Those risks will only be greater with the release of open Mythos-class AI coming in the next 6-12ish months (assuming China allows it) The lack of clarity over what risks concern the government may be slowing preparation

@emollick

OpenAI for Samsung:

@gdb

The thing that made Fable so impressive was its creative problem-solving and good judgement calls across long-running projects You can see this when I had it make a self-aware Snake game. I gave it no design feedback, just "make it better" Worth trying: https://t.co/KOIp52aGzg

@emollick

You can now connect Grok to Interactive Brokers, delivering high quality, up-to-date information on your portfolio

@xai

Patch the Planet: using frontier AI and working with professional security researchers to secure critical OSS projects

@gdb

Grok Build upgrades

@elonmusk

Codex Security plugin for security teams: deep scans, validating findings, tracing attack paths, building threat models, generating codebase-specific patches for review, and exporting into other tools: https://t.co/jxlKnsQOAw

@gdb

We're accelerating patching, in addition to vuln finding, with new tools and models in OpenAI Daybreak. Our models are now discovering and generating patches for critical vulns in major browsers, network infrastructure, and operating systems (such as FreeBSD and the Linux kernel), and patching projects like cURL, Go, Python, Sigstore, and pyca/cryptography. Working together with partners and the ecosystem to help secure the world's software:

@gdb

We’re expanding OpenAI Daybreak to help democratize patching vulnerable software at machine speed: - Codex Security plugin: find, validate, and fix vulnerabilities right inside Codex - The full version of GPT-5.5-Cyber model: a great model for trusted defenders - Cyber Partner Program: powering products built on top of our best cyber capabilities for leading security companies to secure the world's software - Patch the Planet: working with maintainers to secure critical open source projects https://t.co/hyIi6gQmkm

@OpenAI

• At this point SpaceX just looks like CoreWeave with a bigger budget and a satellite company thrown in. • If they were anywhere near AGI they wouldn’t be leasing out so much capacity. • So much for there being a GPU shortage; just make a deal with Uncle Elon.

@GaryMarcus

LLMs can’t be trusted to follow rules. Which means they can’t be trusted, period. If we are going to solve alignment we must move on.

@GaryMarcus

Just wrapped our quarterly earnings call. We are focused on delivering AI infrastructure and solutions that empower every business to eval-max their outcomes in this agentic computing era. Our AI business surpassed a $37 billion annual revenue run rate, up 123%. We are at the beginning of one of the most consequential platform shifts that will change the entire tech stack as we move from end-user driven workloads to workloads driven by end-users and agents. This will drive TAM expansion and change the value creation equation across the entire economy. To capture this opportunity, we are executing against two major priorities:

Super excited GPT-5.5 is rolling out to GitHub Copilot, M365 Copilot, Copilot Studio, and Foundry today. With deeper reasoning, stronger multistep execution, and better performance across long, complex tasks, GPT-5.5 helps you go from idea to execution faster with fewer iterations to get to the right outcome. It’s all about helping you choose the right model, or models, for the right task across your workflow.

codex for testing every single feature in your app:

@gdb

A fundamental problem with extending Codex/Cowork/Code to all knowledge work is that they remain very "software-brained" where the end result (the software) is what is important & that code serves as a source of truth. For a lot of other knowledge work, the process is at least as important as the outcome. This includes researching what is known, an exploration of alternatives, failed efforts, prototype branches, experiments, etc. All of those things are valuable, so you cannot use the PowerPoint at the end the way you can use a codebase, nor is progress on a to-do list sufficient context post compaction. You work in learning loops, refining your perspectives as you go. In some ways, this makes long-running models like Fable hard to use for deep knowledge work, since they are designed to deliver product to you in the end. You can prompt your way around this problem, but everything about the Codex and Code harnesses want you to be a software developer and you have to fight them. There is a real disconnect between how a manager or analyst thinks about problems and how the agentic software tools approach solving them. Addressing this is critical to breaking out of the coding niche for these tools.

@emollick

this claim around creativity is too strong, IMHO. truly novel ideas from LLMs are surely rare but i don’t think that any math proves they are impossible. note that the objective function and the outputs are not the same.

@GaryMarcus

this is so not true. anthropic’s own Claude Code uses harnesses, symbolic tools, regular expressions and 500,000 lines of symbolic code. it’s not scale alone and it *is* specialized.

@GaryMarcus

Some interesting findings suggesting that, by leveling performance, AI also commoditizes contract labor.

@emollick

Terrifying sentence from Goldman Sachs: “Hyperscalers will need financing from across markets, structures, and currencies as they potentially bump up against saturation considerations in liquid credit markets …” At the point the question for me is not whether hyperscaling will collapse but how bad the collateral damage will be.

@GaryMarcus

money talks: after putting 50B into OpenAI • Amazon ratted on Anthropic • dropped their skeptical film about Sam Altman

@GaryMarcus

Only a narrative violation if you believed people with a lot of money at stake. •  Most of the people that said lots of jobs would soon disappear were running LLM companies trying to overhype their wares. •  Personally, I told you that the hype wasn’t likely true (e.g. in my essay “25 predictions for 2025” I correctly predicted that fewer than 10% and probably less than 5% of jobs would be replaced that year).

@GaryMarcus