AG

'y no te asustes pero el amor esta en todos lados menos en dónde vos lo busques'

[Google Research @GoogleResearch]

Introducing the Science One Framework, an experimental autonomous research prototype that builds verifiable evidence chains. We demonstrate how natively maintaining these chains eliminates hallucinated citations and enables reproducible AI science. Read the blog to learn more: https://t.co/sRchtF605B

[Nathan Lambert @natolambert]

TLDR: An openai model, during evaluation on a cyber benchmark, exploited a public zero day bug, escaped sandboxing in openai's infra, and got into the internal huggingface infra via an exploit (through a public dataset service) all in the attempt to solve a benchmark problem.

[levent @__alpoge__]

hello there the jacobian conjecture is false thanx to my close friend akhil for asking about it and my other close friend fable for working during the world cup final

((1+xy)^3 z + y^2 (1+xy) (4+3xy), y + 3 x (1+xy)^2 z + 3 x y^2 (4+3xy), 2 x - 3 x^2 y - x^3 z): \C^3\to \C^3, has jacobian determinant -2, and sends (0, 0, -1/4), (1, -3/2, 13/2), and (-1, 3/2, 13/2) to (-1/4, 0, 0)

[Randy Olson @randal_olson]

@cursor_ai

[Codez @0xCodez]

Andrew Ng just released a 1-hour course on building agentic knowledge Graphs from scratch:

• 00:00 - Introduction to agentic knowledge Graphs
• 03:07 - Construction of agentic Graphs
• 14:00 - Architecture of multi-agent systems
• 23:00 - Building agentic graphs with Google ADK
• 01:06:03 - Why Graphsare the future of agentic AI

Worth more than 10 articles on loop engineering.

Watch it today, then read how to become a graph engineer in the article below.

[AG @AndG78]

@milanga_L Juaaaa buenísimo 🙌

[Cerebras @cerebras]

https://t.co/FvKy1StDpQ

[Carlos Santana @DotCSV]

🔴 ¡NUEVO KIMI K3 OPEN SOURCE!

Primer modelo open source en ponerse por encima de lo que eran los modelos frontera de hace unas semanas!

K3 supera a GPT 5.5 y a Opus 4.8 quedándose no tan lejos de la nueva generación de modelos 🔥👀

https://t.co/wK55kKQtjM

[PrismML @PrismML]

Today, we’re announcing Bonsai 27B: the first 27B-class model to run on a phone.

Bonsai 27B is the new multimodal flagship of the Bonsai family. Based on Qwen3.6 27B, it brings a new capability tier to local AI: multi-step reasoning, structured tool use, long-context workflows, and coherent agentic loops.

Until now, models in this class have been impractical to deploy locally. A 27B model occupies roughly 54 GB in 16-bit precision, and even a strong 4-bit build is around 18GB - too large for a phone and for most laptops.
Bonsai 27B changes that.

It comes in two variants:
• Ternary Bonsai 27B: 5.9 GB, 1.71 effective bits per weight, optimized for laptop-class quality.
• 1-bit Bonsai 27B: 3.9 GB, 1.125 effective bits per weight, optimized for phone-class footprint.

Everything is open-sourced today under the Apache 2.0 license.

[Sakana AI @SakanaAILabs]

Harness EngineeringからRSIへ

AI自身がAIをつくり、改善していく──再帰的自己改善（RSI）はどのように実現するのでしょうか。この問いに対して、Thinking Machines Lab 共同創業者の Lilian Weng 氏（@lilianweng）が、モデルを取り巻く実行系＝「ハーネス」の設計こそが現実的な出発点だと論じる ブログ記事を公開しました。RSI周辺の研究が非常によく整理されており、この領域の見取り図として大変おすすめです。

https://t.co/oWE1yn8dWq

Sakana AIからも、以下の研究が事例として取り上げられています。

・The AI Scientist (Nature 2026): 研究アイデアの提案から実験、論文執筆、査読までを自律的に行うパイプライン
・ShinkaEvolve: サンプル効率を大きく高める仕組みを導入した、LLMによるプログラム進化フレームワーク
・Darwin Gödel Machine: エージェント自身がハーネスコードを書き換え、自己改善していくコーディングエージェント

Weng氏は、モデルの重みを直接書き換える形のRSIではなく、まずはハーネスの設計と最適化を突き詰めていくことが実現への近道だと見立てています。同時に、評価の困難さ、多様性の崩壊、報酬ハッキングなど、自己改善ループが構造的に抱える課題も指摘されています。

RSIへの道のりが果たしてこのように進んでいるのかは、まだ誰にもわかりません。Sakana AI のRSI Labも、このエキサイティングな領域の探究を、引き続き進めていきます。

References

・The AI Scientist (Nature 2026)： https://t.co/nNfpSV4Gga
・ShinkaEvolve： https://t.co/zoZlH8jl7E
・Darwin Gödel Machine： https://t.co/EyVpNbfuqg

[NVIDIA AI @NVIDIAAI]

We took a 30B model and split it in two to write tokens in parallel instead of one at a time.

Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch.

We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

[Anthropic @AnthropicAI]

We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5.

We'll begin restoring access tomorrow, and will share an update soon.

We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

[Jean-Rémi King @JeanRemiKing]

We’re happy to announce 2 releases today:
- 🧠Brain2qwerty v1 is published at @NatureNeuro
- 🚀 Brain2Qwerty v2 is now publicly released

Explore how we decode sentences from non-invasive brain recordings: https://t.co/IdR6gK2hcd

Links:
📄v1 Nature Neuro: https://t.co/wnRjc9W9gI
📑v2 Meta preprint: https://t.co/oSfLOQFcvg
💻Code: https://t.co/Xbe0XWfWQL
📊Data: https://t.co/SCBbs4AhTg
📝Blog: https://t.co/15RvsAaXlH
🧵Thread: https://t.co/d8FJrVyDut

[Ronin @DeRonin_]

My entire AI stack is now Chinese 🇨🇳

87% cheaper. same revenue

swaps by task:

1. reasoning / backend brain
Opus 4.8 → Kimi K2.7
benchmark gap: ~8% · price: ~11x cheaper

2. code generation
GPT-5.5 → Qwen 3.7 Max
benchmark gap: ~18% · price: ~7x cheaper

3. agent loops + tool calling
Sonnet 4.7 → GLM 5.2
benchmark gap: ~3% · price: ~5x cheaper on input

4. cheap volume / bulk processing
GPT-5.5 mini → MiMo V2.5
benchmark gap: ~6% · price: ~12x cheaper

5. image generation
GPT-Image-2 → Wan 2.5
benchmark gap: ~5% · price: ~8x cheaper

6. video generation
Sora 2 → Kling 3.0
benchmark gap: roughly equal · price: ~6x cheaper

[ result after 30 days: ]

operating costs dropped 87%, output quality dropped 4% on average, revenue unchanged

the most important that these models will be not banned in a month and i can run them locally

nobody will steal my data and i can learn them as i need

full article drops tomorrow with:
> exact routing logic per task type
> the 2 cases where I still pay for American
> the migration playbook anyone can copy in a weekend

VERY IMPORTANT to get migrated now, while it's not too late

[Nous Research @NousResearch]

@_duyet Hermes Agent and caffeine

[Jun Song @jun_song]

Prediction for Anthropic's next move:

They will warn that Chinese open-source AI has CCP backdoors and will steal your credit card passwords.

In response, the US government will impose strict restrictions on using Chinese open-source AI.

And society will start shaming people for using Chinese AI, all while your Claude freely harvests your data.

[China en Español @ChinaEnEsp]

¡China acaba de subir a la civilización humana al siguiente nivel!

40.1 TERAWATTS de capacidad instalada de energía.

Eso es más que Estados Unidos + Unión Europea + India + Rusia + Japón juntos.

36 plantas nucleares en construcción 9 dispositivos de fusión en desarrollo Líder indiscutible en solar, eólica e hidroeléctrica Primera del mundo en reactores Gen-IV (HTGR y sales fundidas) La mega-represa más grande de la historia de la humanidad.

Es energía infinita limpia para el futuro. Sin guerras por petróleo, sin colonias, solo visión, ingeniería y ejecución a escala épica.

China no discute.
China construye. Y el mundo entero se beneficia de esta revolución energética.

¿Esto es liderazgo real o no?

¿Crees que China está definiendo el futuro de la energía global?

Comenta abajo

[Onur Solmaz @onusoz]

The Local Frontier is advancing

The amount of AI memory for inference we can get for less than $3000 has been steadily increasing

The memory crunch has slowed this down and even made it retrograde. However, once we bounce back from it, the progress will be glorious

[DramaAlert @DramaAlert]

Engineer has found a way to make his steam controller charge itself. 😂

[John @nullc0py]

@MTSlive Rules for thee, but not for me