Ascend
What's new in AI

What's new in AI: 08 Nov 2026

08 Nov 2026 · All digests

Key developments include major funding for OpenAI, Anthropic's accelerated AI development, export concerns over Nvidia chips, exposed AI hacking tools, and a new robot learning benchmark.

OpenAI reportedly raises an additional $11 billion at a $1.25 trillion valuation

OpenAI announced a fresh $11 billion funding round that pushes its valuation to $1.25 trillion, one of the highest for an AI company.

The influx of capital will accelerate OpenAI's product rollout and influence market dynamics for AI services.

Source: TechCrunch AI

Why a new report on Anthropic automating AI development is alarming researchers

Anthropic disclosed that it uses AI to design and improve its own models, achieving roughly twice the speed of progress compared to a year ago.

Automating AI creation could dramatically speed up capability growth, raising safety and alignment concerns.

Source: The Verge AI

The US let Nvidia sell chips to an empty office in China

Reports indicate Nvidia's H200 AI chips were shipped to a Chinese entity with minimal vetting, sparking criticism of export-control oversight.

Weak oversight can enable advanced AI hardware to reach jurisdictions with fewer safeguards, affecting global AI governance.

Source: The Verge AI

Hundreds of Artex AI hacking tools found publicly exposed to the web

Security researchers discovered dozens of servers openly hosting the Artex penetration-testing suite, giving anyone internet access to powerful AI hacking tools.

Easy access to AI-driven hacking utilities raises the threat level for cyber attacks across industries.

Source: TechCrunch AI

ToyBlocks: A controlled world for robot learning and evaluation

AllenAI released ToyBlocks, a simulated environment that standardizes robot learning tasks and provides benchmarks for embodied AI research.

A common benchmark helps researchers compare approaches, accelerating progress in robot perception and control.

Source: Hugging Face

What to learn from this

Turn today's news into a plan

Study AI model safety and evaluation techniques, focusing on automated model generation and benchmarking tools like ToyBlocks. Understanding these areas will help you assess rapid capability gains and protect systems from emerging AI-powered threats.

Build my AI Engineer plan