What is fascinating about China's approach to robotics is how diverse it is. Lots of local organizations and parties putting up their own research centers like this one: www.globaltimes.cn/page/202504/...

Whereas in the US, things seem much more concentrated

Introducing ASAL: Automating the Search for Artificial Life with Foundation Models

Blog: sakana.ai/asal/

We propose a new method called Automated Search for Artificial Life (ASAL) which uses foundation models to automate the discovery of the most interesting and open-ended artificial lifeforms!

Looking Back At 2024 AT Protocol Development | Bluesky In May 2024, we published a 2024 Protocol Roadmap, and we want to give an end-of-year update. We will follow up soon with a forward-looking roadmap for 2025.

docs.bsky.app/blog/looking...

bsky.social/about/blog/1...

On the Measure of Intelligence To make deliberate progress towards more intelligent and more human-like artificial systems, we need to be following an appropriate feedback signal: we need to be able to define and evaluate intellige...

arxiv.org/abs/1911.01547

Surreal how this work has become even more pertinent now in 2025. Still criminally underrated in the mainstream ai community

Let's go 🌪️🚀

Building your own algorithms for social networks - Bluesky Custom Feeds with SkyFeed Most social networks use proprietary algorithms to recommend new content and keep you on their platform for as long as possible. Other pl...

media.ccc.de/v/37c3-light...

A look at the future: chatting with my robot via Discord to ask it to find something in my house.

This uses an LLM to understand what the human wants and generate a task plan, then builds an open-vocab 3d scene representation to find and pick up objects

the fact that openai finally acknowledged it for their o3 marketing (and now suddenly everyone is talking about it) is a testament to it being fundamental to humanity building "agi"

I believe the insinuation with calling arc-agi's benchmarks as goalposts is a massive disservice to its mission. Their philosophy and vision is still the same with V2 if you knew them before o3

noway they could do something like that. Their ethics and values would never allow them to do that. We are safe under our overlords they would never let any harm come to us 🙏

Prophecies of the Flood What to make of the statements of the AI labs?

There has been a definite shift in recent weeks where insiders in the various AI labs are suggesting that very intelligent AIs are coming very soon.

I wrote a bit about why this might be happening and what we can take away from their apparent confidence. www.oneusefulthing.org/p/prophecies...

This is the promise of bluesky ✊ On x, I felt it too, the recent right wing conservative push is quite evident actually. Kind of obvious tho given the political climate. But this is also why we need auditable algorithmic feeds more than ever now.