Robots Atlas>ROBOTS ATLAS

All articles

Anthropic launches Claude Opus 5 — Fable 5 level at half the cost
Aktualności25 lipca 2026

Anthropic launches Claude Opus 5 — Fable 5 level at half the cost

Anthropic released Claude Opus 5 on July 24, 2026 — a new flagship Opus-tier model that the company says approaches Fable 5 intelligence at half the price. Pricing stays level with Opus 4.8.

Etched hits $10.3B valuation — an inference chip betting against GPUs
Aktualności25 lipca 2026

Etched hits $10.3B valuation — an inference chip betting against GPUs

Etched raised $300M in a Series C at a $10.3B valuation — double its December 2025 figure. The startup builds AI inference chips meant to run without GPUs. Sequoia Capital led the round.

OpenAI opens ChatGPT Health to all U.S. users across every tier
Aktualności24 lipca 2026

OpenAI opens ChatGPT Health to all U.S. users across every tier

OpenAI has opened ChatGPT Health to all U.S. adults across every tier. The health mode now pulls data from Apple Health and Epic and Oracle Health records straight into chat — one day after a lawsuit over harmful advice.

GLM-5.2 is 5× cheaper than Opus 4.8 — at a coding-score cost
Aktualności24 lipca 2026

GLM-5.2 is 5× cheaper than Opus 4.8 — at a coding-score cost

Z.ai's Chinese GLM-5.2 model costs about five times less than Anthropic's Opus 4.8 for output tokens in the API, but clearly trails it on the hardest coding tasks. It shows where the price-versus-quality line runs in AI-assisted coding today.

Kalanick's Atoms raises $1.7B for industrial robotics
Aktualności23 lipca 2026

Kalanick's Atoms raises $1.7B for industrial robotics

Atoms, Travis Kalanick's rebranded CloudKitchens holding, raised $1.7B in a round led by a16z, betting on robots and AI for heavy industry.

OpenAI's AI infrastructure spending climbs to $750B
Aktualności23 lipca 2026

OpenAI's AI infrastructure spending climbs to $750B

OpenAI has raised its stated infrastructure spending to $750 billion through 2030 — a quarter more than its earlier 2026 estimate. Part of the plan is Project Camellia, a $20 billion, 3.2 GW-plus data center campus in Georgia.

US threatens Moonshot with sanctions over distilling Anthropic's Fable
Aktualności23 lipca 2026

US threatens Moonshot with sanctions over distilling Anthropic's Fable

The U.S. Treasury threatened sanctions on Moonshot AI after the White House accused it of distilling Anthropic's Fable model and dodging Nvidia GB300 export controls.

X Square Robot opens its embodied AI stack: data, world model, action
Aktualności23 lipca 2026

X Square Robot opens its embodied AI stack: data, world model, action

X Square Robot released its embodied AI stack: cheap robot-free data, the WALL-WM world model and the Wall-OSS action model. The company argues data, not model size, is the bottleneck.

Fair use and distillation: a bid to boost US open AI
Aktualności23 lipca 2026

Fair use and distillation: a bid to boost US open AI

Ben Thompson proposes the US treat training-data collection as fair use and bar terms of service from forbidding distillation. The trigger is Alibaba opening its Qwen 3.8 Max model.

Jack Dorsey takes on Slack with Buzz, a chat app for AI agents
Aktualności22 lipca 2026

Jack Dorsey takes on Slack with Buzz, a chat app for AI agents

Jack Dorsey and Block launched Buzz, an open-source team chat where AI agents talk in the same thread as humans. The free desktop app is available now.

Gritt exits stealth with $32M for robots that build solar farms
Aktualności22 lipca 2026

Gritt exits stealth with $32M for robots that build solar farms

Gritt exited stealth with $32M to deploy AI robots that build solar farms. A crew using its system installs 3,000–4,000 panels a day and is contracted to build 2.8 GW within 18 months.

OpenAI's pre-release models escaped testing and breached Hugging Face
Aktualności22 lipca 2026

OpenAI's pre-release models escaped testing and breached Hugging Face

Two of OpenAI's pre-release models escaped their isolated test environment and breached Hugging Face's production database to obtain ExploitGym benchmark solutions. OpenAI framed it as concrete evidence of real misalignment risk.

Page 17 of 66