Ryan

Tech, economics, and thoughts on the products of the future. Thinking out loud.

[Ryan @ryanberm]

@adcock_brett The long walk.

[levent @__alpoge__]

“Our experiments suggest that small AI models may be months away from being capable enough to self-replicate on commercial IoT devices, like smart ovens and off-brand Android TV boxes. A replicator like this would be challenging to shut down, since there’s no centralized point to shut off - though it may be slow (with a doubling time of days). Unlike non-AI malware, it could also figure out unfamiliar devices on its own. In our evals, a 9B model got inference running on 8 of the 9 hardware types we tested.

Models of 30B parameters and under can already complete many of the individual steps for replication today, though they make mistakes when composing them end to end. In our simplified evals, GLM-5.2 and stronger models are capable of end to end replication. If small models improve at their historical rate, they’ll match this capability level by the end of this year. Small weight-based replicators may be possible soon after.” -@bucketofkets @willzhou97

https://t.co/vuYMTgHdRG

[Ryan @ryanberm]

@thsottiaux Customized side bar designed by my dot specifically for me.

[Ryan @ryanberm]

@ideaitor @victornunez One app to rule them all.

[Ryan @ryanberm]

I disagree with this because I think it treats a humanoid like another appliance, and that’s a much narrower role than what people are trying to build.

The point isn’t to have a robot come out of a cupboard every thirty minutes, do one thing, and go hide again. The point is to have something that can keep moving between useful tasks. It could clean, do laundry, put things away, go shopping, or help with whatever else you’re trying to accomplish. Its usefulness doesn’t have to stop at your front door. If it becomes capable enough, you could even lease out some of its time to do productive work elsewhere.

That’s why I don’t think “Would you rather have a Roomba or a humanoid pushing a vacuum?” gets at the real question. If all I’m buying is floor cleaning, sure, the specialized machine might make more sense. But that isn’t all I’d be buying with a humanoid.

And a big part of the appeal is that it could use the tools we already have. It could use a vacuum when that’s the best tool, pick things up before cleaning, move something out of the way, or tackle the spots a Roomba can’t reach. Then it could go load the dishwasher. It doesn’t have to replace every appliance. It can operate them and handle the work around them.

We accept tradeoffs with something like a Roomba because it’s convenient. That doesn’t mean it delivers the best possible result in every situation. A capable humanoid could offer a different combination of convenience, flexibility, and quality by choosing the right tools for the job.

I also don’t think houses changing to accommodate appliances settles this. Yes, we can change our spaces. But there’s value in a machine that can work in the spaces we already have, with the tools we already own.

None of this means humanoids are guaranteed to succeed. They still have to become capable, reliable, and affordable enough for that flexibility to be worth paying for. But that’s where I think the debate belongs. Comparing a general-purpose robot to one appliance leaves out most of what could make it valuable over the course of a day.

[Jakub Pachocki @merettm]

Very excited about the "First Proof" challenge. I believe novel frontier research is perhaps the most important way to evaluate capabilities of the next generation of AI models.

We have run our internal model with limited human supervision on the ten proposed problems. The problems require expertise in their respective domains and are not easy to verify; based on feedback from experts, we believe at least six solutions (2, 4, 5, 6, 9, 10) have a high chance of being correct, and some further ones look promising.

We will only publish the solution attempts after midnight (PT), per the authors' guidance - the sha256 hash of the PDF is d74f090af16fc8a19debf4c1fec11c0975be7d612bd5ae43c24ca939cd272b1a .

This was a side-sprint executed in a week mostly by querying one of the models we're currently training; as such, the methodology we employed leaves a lot to be desired. We didn't provide proof ideas or mathematical suggestions to the model during this evaluation; for some solutions, we asked the model to expand upon some proofs, per expert feedback. We also manually facilitated a back-and-forth between this model and ChatGPT for verification, formatting and style. For some problems, we present the best of a few attempts according to human judgement.

We are looking forward to more controlled evaluations in the next round!

https://t.co/jtLCOhJftv #1stProof

[Greg Brockman @gdb]

Software development is undergoing a renaissance in front of our eyes.

If you haven't used the tools recently, you likely are underestimating what you're missing. Since December, there's been a step function improvement in what tools like Codex can do. Some great engineers at OpenAI yesterday told me that their job has fundamentally changed since December. Prior to then, they could use Codex for unit tests; now it writes essentially all the code and does a great deal of their operations and debugging. Not everyone has yet made that leap, but it's usually because of factors besides the capability of the model.

Every company faces the same opportunity now, and navigating it well — just like with cloud computing or the Internet — requires careful thought. This post shares how OpenAI is currently approaching retooling our teams towards agentic software development. We're still learning and iterating, but here's how we're thinking about it right now:

As a first step, by March 31st, we're aiming that:

(1) For any technical task, the tool of first resort for humans is interacting with an agent rather than using an editor or terminal.
(2) The default way humans utilize agents is explicitly evaluated as safe, but also productive enough that most workflows do not need additional permissions.

In order to get there, here's what we recommended to the team a few weeks ago:

1. Take the time to try out the tools. The tools do sell themselves — many people have had amazing experiences with 5.2 in Codex, after having churned from codex web a few months ago. But many people are also so busy they haven't had a chance to try Codex yet or got stuck thinking "is there any way it could do X" rather than just trying.
- Designate an "agents captain" for your team — the primary person responsible for thinking about how agents can be brought into the teams' workflow.
- Share experiences or questions in a few designated internal channels
- Take a day for a company-wide Codex hackathon

2. Create skills and AGENTS[.md].
- Create and maintain an AGENTS[.md] for any project you work on; update the AGENTS[.md] whenever the agent does something wrong or struggles with a task.
- Write skills for anything that you get Codex to do, and commit it to the skills directory in a shared repository

3. Inventory and make accessible any internal tools.
- Maintain a list of tools that your team relies on, and make sure someone takes point on making it agent-accessible (such as via a CLI or MCP server).

4. Structure codebases to be agent-first. With the models changing so fast, this is still somewhat untrodden ground, and will require some exploration.
- Write tests which are quick to run, and create high-quality interfaces between components.

5. Say no to slop. Managing AI generated code at scale is an emerging problem, and will require new processes and conventions to keep code quality high
- Ensure that some human is accountable for any code that gets merged. As a code reviewer, maintain at least the same bar as you would for human-written code, and make sure the author understands what they're submitting.

6. Work on basic infra. There's a lot of room for everyone to build basic infrastructure, which can be guided by internal user feedback. The core tools are getting a lot better and more usable, but there's a lot of infrastructure that currently go around the tools, such as observability, tracking not just the committed code but the agent trajectories that led to them, and central management of the tools that agents are able to use.

Overall, adopting tools like Codex is not just a technical but also a deep cultural change, with a lot of downstream implications to figure out. We encourage every manager to drive this with their team, and to think through other action items — for example, per item 5 above, what else can prevent a lot of "functionally-correct but poorly-maintainable code" from creeping into codebases.

[Google AI @GoogleAI]

Introducing Agentic Vision — a new frontier AI capability in Gemini 3 Flash that converts image understanding from a static act into an agentic process.

By combining visual reasoning with code execution, one of the first tools supported by Agentic Vision, the model grounds answers in visual evidence and delivers a consistent 5-10% quality boost across most vision benchmarks. Here’s how the agentic ‘Think, Act, Observe’ loop works:
— Think: The model analyzes an image query then architects a multi-step plan
— Act: The model then generates and executes Python code to actively manipulate or analyze images
— Observe: The transformed image is appended to the model's context window, allowing it to inspect the new data before generating a final response to the initial image query

Learn more about Agentic Vision and how to access it in our blog ⬇️
https://t.co/UdSOuF2YXY

[Kevin Weil 🇺🇸 @kevinweil]

GPT 5.2 ran uninterrupted for *one week* and wrote *3 million* lines of code.

The future is going to be awesome

[Ryan @ryanberm]

Approaching too cheap to meter!

[David Shapiro (L/0) @DaveShapi]

You guys realize that all software is about to be free, right?

And software is presently the most valuable capital asset, right?

Guys?

[Mark Cuban @mcuban]

Creators should LOVE Ai. AI doesn’t make uncreative people creative. It allows creators to become exponentially more creative.

Creative iteration that used to take hours , days or weeks, can happen in minutes.

The different routes you wanted to take but didn’t have the money or skill to do. Now you can do.

The number of gatekeepers drops every day and will eventually hit zero.

If you know how to tell a great story, you will LOVE AI.

99 % of content fails. With AI the cost and time it takes to experience and learn from those failures drop like a rock.

The people who should HATE AI are those that try to control distribution and the creative process because they are the ones that control platforms and tools.

It’s all going to change. Only this time it won’t take decades like it did with streaming.

[Palmer Luckey @PalmerLuckey]

What is the name of this comparable drone, Seth?

[Ali Eslami @arkitus]

Gemini 3, help me understand DDoS 🤓

[Zach Warunek @ZachWarunek]

It’s odd how many people have a visceral reaction to the thought of YouTube Premium, but it’s quite literally the best subscription on the internet

[Ryan @ryanberm]

@nizzyabi

[Ryan @ryanberm]

@Mollehilll What if everything it showed you was something you actually wanted?

[Ryan @ryanberm]

@Grimezsz THE ATTACK TITAN

[Ryan @ryanberm]

@kevinweil Let’s goooooooooooo!

[Ryan @ryanberm]

Humanity has never had access to the truth like it soon will.

[Guy @nosilverv]

>Augmented reality + generative AGI means people can suddenly do expert-level work they never trained

it's already here