Stripe acquires OpenRouter for $7 billion. The AI agent economy is gaining momentum

The payment giant has bought the AI model router for a staggering sum, and there's a clear explanation behind it.

Stripe has been building infrastructure for the AI agent economy for several years now.

Already today, AI agents consume more tokens than humans, and a significant volume of transactions is already flowing through payments.

Which raises a simple question: who's actually processing the payment at that moment? Stripe, of course.

Physics of Agents

As AI agents become more prevalent, they'll interact and influence one another.

Can we predict the collective behavior that emerges?

Surprisingly, their dynamics follow compact, predictive laws of statistical physics

Researchers studied >10,000 communities of LLM agents that talk to each other and update opinions over time.

Team show that collective behavior of interacting AI agents can be modeled and predicted using stat mech.

The key insight is that AI agents drift to low energy states that minimize their social pressure.

Understanding such collective behavior is important for safety and alignment of multi-agent systems.

Meet Q-Learning with World Models (QWM)

World models have emerged as one of the biggest directions in physical AI.

At the same time, RL fine-tuning is unlocking capabilities in frontier models beyond what pretraining can achieve on its own.

Can we get the best of both worlds?

Meet Q-Learning with World Models (QWM).

QWM isn’t tied to a specific RL algorithm it’s a general framework for test-time scaling in RL.

On top of RLPD, it also shows consistent gains.

QWM scales to frontier video models.

QWM with frontier video models also consistently improves over the base RL method

New stealth model just dropped: Ox Alpha

A mysterious frontier model appeared on OpenRouter and OpenCode under the Stealth provider no company name, no official claim.

Try it while it’s free.

Key specs:
• 1,048,576-token context window
• Max output: 131k tokens
• Multimodal: text + image + video input
• Built for coding, long-horizon agentic work, and production workloads
• Tool calling supported
• Completely free for one week with generous (near-unlimited) rate limits
• Provider claims zero training on your prompts/completions during the test period.

Community fingerprinting is already pointing strongly toward a multimodal variant of Z.ai’s GLM-5.3 family (tokenizer matches almost perfectly, same API signature, reasoning modes, etc.).

Xiaomi’s MiMo team is the other main speculation, but the evidence currently leans GLM.

Early users are calling it genuinely frontier-level for coding and sustained agentic tasks fast, strong reasoning, and surprisingly polished.

No official benchmarks yet, but the real-world feedback is already heating up.

Waymo introduced their first custom silicon

Waymo collaborate with industry leaders including AMD, Micron, NVIDIA, Samsung, Sandisk, Socionext, and TSMC to build the most capable computing system on the road.

Claude Academy is now live.

Whether you're figuring out what AI is or already using Claude every day, there's a path that meets you where you are. The courses and tutorials are free and open to anyone at academy.claude.com

Hugging Face, an AI developer platform, has been exploring a potential $13 billion sale, highlighting its key role in the open-source AI ecosystem.

One of the most interesting things happening in open AI right now

Marin Community is currently training a massive MoE model:
535B total parameters (23B active)
on 18+ trillion tokens
And everything is fully public.

You can watch in real time: • training loss
• scaling ladder
• configs
• data mixture

The first CPU built for agents is going to work at scale.

SpaceX is deploying NVIDIA Vera to accelerate the orchestration, code execution, and data processing that powers its next generation of agentic AI keeping GPUs fed and agents acting fast.

From gigawatt AI factories to orbit. One NVIDIA architecture, everywhere

Cool idea. There are a lot of interesting harness designs that are starting to emerge around tool calling and code execution. RLM is one of them.

But so is this Speculative Programmatic Tool Calling approach (from the same author of RLM).

A general class of technique for speculating on tool calls during code generation in a harness and queuing them early to overlap with token generation + REPL execution time.

Google revolutionized environment design by building the harness for environments, not just agents

Static environments bottleneck the growth of LLM agents.

When environments can't evolve, agent progress stalls.

Meet EnvHarness: a powerful, unified wrapper framework engineered to dynamically adapt environments and unlock rich training signals:

- Non-Intrusive Interception
- Dynamic Environments
- Plug-and-Play Integration

GitHub
Paper

Hot Chips '26: OpenAI says its Gen 1 chip is only step one with Gen 2 nearing tape-out in a number of months and Gen 3 already operational, afterwards bringing out the Broadcom and Celestica teams to share the stage

"One more thing this is step one of a multi-generational roadmap."

"We have Generation 2 already well under development, and we're heading towards tape out in some number of months, and we have Gen 3 already operational…"

"The whole point is that this is going to lower the cost of infrastructure, as I said at the beginning."

"And I also want to give a really big shout-out to our partners, Broadcom and Celestica, who are very key partners for us to be able to deliver this capability."

Lookin for an AI-native SDLC playbook? Anthropic published a detailed one

It keeps the standard SDLC stages (plan/design/build/test/deploy/maintain) but reframes the process from a linear flow to artifact-driven loop.

Interesting read.

Google introduced Gemini 3.5 Transcribe, a new speech to text model with smart transcription, function calling, more precise transcription (lower WER), custom vocabulary support, multi-speaker identification, and support for over 85 languages!

Also with realtime streaming support.

Hugging Face have a huge news today

1. Nvidia has agreed to buy Hugging Face for $12.9 billion, The Information reported.

2. Today Hugging Face presented the first truly accessible RL robot Microduck

A 25 cm tiny open-source biped with 15 actuators and packed with sensors (camera, speaker, LiDAR, NFC, bluetooth, wifi, etc) that you train yourself with reinforcement learning.

It's also playable out of the box with more than half a dozen fun and playful pre-trained policies to have it walk, sit, crouch, roller-skate, pick up objects with its articulated beak, and recover on its own.

And all for less than $400.

Anthropic kicking off the first phase of the research preview for Model Hardware Standard (MHS): a new standard for AI agents to safely operate physical equipment in scientific research and advanced manufacturing.

Connecting AI to hardware requires days or weeks of bespoke integration, with no standard way for agents to operate equipment safely.

MHS cuts integration to hours or minutes, provides an interface that makes devices discoverable, and enables agents to operate them safely.

In early testing, AI agents used MHS to:

Run a drug-discovery experiment with real-time error handling at Genentech

Compress an imaging experiment from weeks to a day at HHMI Janelia Research Campus

Improve laser stabilization on QuEra's quantum computers from 58% to 99.3%

MHS currently best covers lab and manufacturing equipment. Many developers are already using Claude Code to operate hardware like boards and cameras; Anthropic’s research preview will help extend MHS to these devices, so they can all work under one interface.

Google introduced GlucoFM, a lightweight, self-supervised continuous glucose monitoring foundation model

That separates metabolic baselines from transient spikes, producing transferable representations and setting new performance standards across diverse metabolic prediction tasks, such as diabetes risk assessment, insulin resistance, and post-prandial glycemic response.

a16z is announced the Machine Age Fund, a new $1.1 billion fund for founders rebuilding what intelligence runs on: chips, memory, networking, systems software, power, and the machines that bring AI into the physical world.

New Anthropic’s research: Can Claude autonomously align other AIs?

Researchers gave Claude 48 hours and 1 GPU to improve the alignment of small models.

It researched and proposed methods, then trained and tested the models on its own. It worked surprisingly well.

Could a model one day align its stronger successors?

As a first test, team had Sonnet 5 post-train an early checkpoint of Opus 4.8, a more capable model.

It reached safety scores approaching those of production Opus 4.8, which went through full alignment training.

Claude can reliably fix measurable misalignment. But subtle or rare failures may have no benchmark at all so everything hinges on measuring the right things.

OpenAI has quietly started offering major customers outcome-based pricing: they only pay when its AI completes tasks like customer-support interactions.

This is the beginning of the shift from selling tokens to selling completed work. Big Move before GPT-Astra Release.