AI this week: Sep 14 – 20, 2026
2 new models, 0 price changes and 109 articles from the sources we follow.
Top stories
Emerald AI, Google and NVIDIA Launch Alliance to Advance Flexible AI Data Centers
AI factories are the infrastructure of the intelligence era. Scaling them responsibly will depend as much on innovation across the grid as inside the data center. Today, Emerald AI, Google and NVIDIA announced the…
MilleMiglia: A realistic instance generator for middle-mile logistics
Algorithms & Theory
New experts join Google’s AI & Economy team
We are expanding our AI & Economy team with world-class academic advisors, fellows, and core internal researchers.
Co-creating the future of fashion with Google
Google worked side-by-side with designers Jane Wade and Sergio Hudson to custom-design Google Flow tools to prep for NYFW.
Introducing the Australian Youth Safety Blueprint
OpenAI introduces the Australian Youth Safety Blueprint, a six-pillar roadmap for safer AI experiences that protect and empower young people.
Partnering with Accenture on embedded evaluation
Partnering with Accenture on embedded evaluation
How Cooley is accelerating IPO work with ChatGPT
Cooley built GO Public with ChatGPT Work to bring intelligence to the IPO process, helping lawyers surface issues earlier and focus judgment where it matters most.
Así regula China la IA frente a EEUU: de límites contra la guerra cognitiva e ideológica a retirar agentes IA peligrosos
La carrera por liderar y armarse contra potencias extranjeras sigue abierta, pero mientras China ha establecido ciertas normativas más que EEUU. Más información: OpenAI, Google y Mastercard firman un llamamiento para…
llm-keys-ui 0.1
Release: llm-keys-ui 0.1 This plugin solves a very specific problem. I've started using Codex Remote to run coding agents on various machines while controlling them from my phone. Sometimes I use those machines to hack…
El ‘freno’ a la IA amenaza al salvavidas billonario de la economía mundial tras los aranceles y la guerra de Trump en Irán
Una economía "colgada de un chip" hace que la potencial ralentización de las inversiones tecnológicas, que este año podrían superar el billón de dólares, sea un asunto clave para la estabilidad financiera global El…
DeepSeek ramène son cache d'attention à 890 octets par jeton, le poste de coût devenu critique pour les agents
DeepSeek a publié le 17 septembre 2026 le rapport technique de V4.1-Flash, modèle multimodal à mélange d'experts de 552 milliards de paramètres servi depuis le 10 septembre, avec un contexte d'un million de jetons. Son…
Harvey lève 550 millions de dollars et rachète Guardrails AI, l'éditeur du cadre open source de contrôle des agents
Harvey lève 550 M$ à 15,5 Md$ de valorisation et rachète Guardrails AI. Avec Tenet (modèle post-entraîné sur Kimi K3), le banc d'essai LAB et désormais les garde-fous et la simulation d'agents, l'éditeur d'IA juridique…
New models this week
| Model | Inputi | Outputi | Contexti |
|---|---|---|---|
GLM 5.3 FlashXZ.ai (Zhipu) | $0.37 | $1.25 | 1M |
Ternary Bonsai 2 27BPrism Ml | $0.075 | $0.50 | 256K |