Sebastian Uribe

AI Sr. PM + Book Author AI for science + animal rights advocate

[Sakana AI @SakanaAILabs]

Introducing The AI Scientist: The world’s first AI system for automating scientific research and open-ended discovery!

https://t.co/jC7g5GPVsE

From ideation, writing code, running experiments and summarizing results, to writing entire papers and conducting peer-review, The AI Scientist opens a new era of AI-driven scientific research and accelerated discovery.

Here are 4 example Machine Learning research papers generated by The AI Scientist.

We published our report, The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery, and open-sourced our project!

Paper: https://t.co/lTQ8UenFHk
GitHub: https://t.co/Im53whVeAq

Our system leverages LLMs to propose and implement new research directions. Here, we first apply The AI Scientist to conduct Machine Learning research. Crucially, our system is capable of executing the entire ML research lifecycle: from inventing research ideas and experiments, writing code, to executing experiments on GPUs and gathering results. It can also write an entire scientific paper, explaining, visualizing and contextualizing the results.

Furthermore, while an LLM author writes entire research papers, another LLM reviewer critiques resulting manuscripts to provide feedback to improve the work, and also to select the most promising ideas to further develop in the next iteration cycle, leading to continual, open-ended discoveries, thus emulating the human scientific community. As a proof of concept, our system produced papers with novel contributions in ML research domains such language modeling, Diffusion and Grokking.

We (@_chris_lu_, @RobertTLange, @hardmaru) proudly collaborated with the @UniOfOxford (@j_foerst, @FLAIR_Ox) and @UBC (@cong_ml, @jeffclune) on this exciting project.

[Eben Bayer @ebenbayer]

I’d like to see some wild bio inspired stratospheric injection solutions.

algae with helium organoidsthat float in the stratosphere multiplying and blocking excess radiation.

Good overlap with biological cloud seeding here for ice formation

[Niko McCarty. @NikoMcCarty]

How can we condense 50-100 years of biological progress into 5-10 years?

I wrote a response to @DarioAmodei's essay, "Machines of Loving Grace." Most bottlenecks slowing biology today, I argue, are biophysical rather than computational.

LEVERS FOR BIOLOGICAL PROGRESS🔻

[Alexander Lavin @AlexLavin_C137]

https://t.co/hmoKe7jCDg "heliolab" showcase happening now!
Physics-AI applications in the @NASA Digital Twin Earth portfolio, led by partners @GoogleAI @NVIDIAAI @SimAI4Science (https://t.co/5desx2LA64)

[Pasteur Labs & Inst. for Simulation Intelligence @SimAI4Science]

+1

[Nantas Nardelli @nntsn]

I'm so *wildly* excited to finally be able to share that I have joined Pacific Fusion and I'm starting an AI team!

PF is the product of a rare alignment of a bunch of exceptional events. The result? A realistic path to fusion energy in years instead of decades.

Story time! 🧵

[Ben Moseley @benm_oseley]

🚀 Excited to share that I’m joining @ImperialCollege in Nov as a Lecturer, leading the Scalable SciML Lab in @ESEImperial! I’m also honoured to be a Schmidt AI in Science Fellow at @ImperialX_AI.

I'm hiring PhD students! https://t.co/H0RmNBJcnL 🌍✨

[Prof. Stefan Rahmstorf 🌏 🦣 @rahmstorf]

44 scientists have issued a stark warning about the threat of a major ocean circulation change, causing a detrimental disruption of climate patterns lasting for centuries to come.
Check out this short video explaining what this is all about.

[ARC Prize @arcprize]

[Paper] Dreamcoder's inductive program synthesis has inspired many ARC-AGI approaches.

By combining neural networks + symbolic abstractions, it can tackle tasks from programming to physics.

https://t.co/IAHsbGsfql

[Sentient @Sentient_Media]

Advances in AI are paving the way for two-way communication with animals, with profound ethical implications. @stevenrouk explains:
https://t.co/IdMq1bGlsg

[Steven Rouk @stevenrouk]

Could AI help us communicate with other animals? If so, what would that mean for animal rights?

Check out my new piece in @Sentient_Media.

#animalrights #AI #artificalintelligence

[Peter Singer @PeterSinger]

I’m excited to announce my new book Consider the Turkey, which reveals some surprising details about the lives of the birds who end up on Thanksgiving tables across America. Check out this short book to inform yourself on what you are eating, including some plant-based recipe ideas.

https://t.co/lqHGIynvHF

[Peter Singer @PeterSinger]

I recently wrote for @TIME on why the U.S. is so far behind Europe in animal welfare standards, and what this means for hundreds of millions of animals. It’s a wake-up call to Americans who continue to buy eggs from hens living their entire lives crammed into tiny cages. I also hope it will draw attention to my just-published book, Consider the Turkey, which is time to stir up some conversation about the treatment of the bird that appears on so many American Thanksgiving tables. Read it here: https://t.co/7kbr6uSaOq

[Ben Reinhardt @Ben_Reinhardt]

1/ For the past year, we’ve been building a new research organization:

Speculative Technologies (@spec__tech) exists to create an abundant, wonder-filled future by unlocking powerful materials and manufacturing technologies that don’t have a home in other institutions.

[CLS @ChengleiSi]

Automating AI research is exciting! But can LLMs actually produce novel, expert-level research ideas?

After a year-long study, we obtained the first statistically significant conclusion: LLM-generated ideas are more novel than ideas written by expert human researchers.

[Nathan Labenz @labenz]

"Evolution doesn't make fixed solutions to specific environments – it makes problem-solving agents at all scales

In AI, the higher level is intelligent, but the parts underneath are dumb

In biology, every layer tries to bend/shape/coerce/hack the layers below & above it

For multi-scale biological kinds of AIs, we'd have to give up on isolating the layers so that each layer can count on the bottom layer being correct"

-@drmichaellevin on lessons from biology for ML architecture design

Dare I say ... self-recommending? 🔗👇

[ARC Prize @arcprize]

ICYMI - @fchollet's AGI-24 conference talk filmed in high-quality by @MLStreetTalk.

MLST interview episode coming soon.

https://t.co/Fng0B9lm6k

[Nicholas Larus-Stone @nlarusstone]

Tired of wasting hours on data analysis instead of actual science? Scientists, we feel your pain.

We’re releasing Metis, an intelligent scientific assistant that helps you go from data to publication-quality figures in under a minute. And it’s free for academics!

The best part? You don’t need to learn a new tool — just tell Metis what you want to do.

Here’s using Metis in the @sphinx_bio platform to fit and plot a dose response curve using plain English!

[Tony Kulesa @kulesatony]

After today's Nobel, there couldn't be a better day to announce this 📢

Together with ARIA, we are building a new fellowship to bring top AI people into cutting edge science labs.

We'll launch next year. Join us Nov 5 in London or reach out online to learn more.

[Asimov Press @AsimovPress]

Today's science fiction story is by @owl_poster.

Read it here: https://t.co/Ar9Cy8PFzN