Z.ai Launches GLM-5.2 With a Usable 1M-Token Context, Two Thinking-Effort Levels, and No Benchmarks at Launch
LLM

Z.ai Launches GLM-5.2 With a Usable 1M-Token Context, Two Thinking-Effort Levels, and No Benchmarks at Launch

Z.ai launched GLM-5.2 on June 13, 2026, featuring a 1-million-token context window and two thinking-effort levels...

MarkTechPost
Orbio raises $21 million to automate hiring and onboarding for frontline workers
Startups & Funding

Orbio raises $21 million to automate hiring and onboarding for frontline workers

Orbio announced a $21 million Series A funding round led by Dawn Capital to support its platform that automates hiring...

TechCrunch Startups
Claude Code Guide 2026: 25 Features with Examples + Demo
Coding Co-pilotAnthropic

Claude Code Guide 2026: 25 Features with Examples + Demo

This guide covers Claude Code, an agentic coding tool with 25 features including CLAUDE.md, skills, subagents, hooks,...

MarkTechPost
A Coding Hands-On on FineWeb for Streaming, Filtering, Deduplication, Tokenization, and Large-Scale Web Corpus Analytics
Research & Papers

A Coding Hands-On on FineWeb for Streaming, Filtering, Deduplication, Tokenization, and Large-Scale Web Corpus Analytics

This tutorial provides a hands-on guide to working with the FineWeb dataset, demonstrating how to stream and analyze...

MarkTechPost
China may have accessed Mythos
Ethics & RegulationAnthropic

China may have accessed Mythos

The White House imposed export restrictions on Anthropic's Mythos AI model due to concerns that it may have been...

The Verge AI
Startups & FundingOpenAI

Introducing the OpenAI Partner Network

OpenAI has launched the Partner Network with a $150M investment to help global partners accelerate enterprise AI...

OpenAI Blog
As AI companies race to go public, who else is along for the ride?
Startups & Funding

As AI companies race to go public, who else is along for the ride?

AI startups are attempting to capitalize on the momentum of high-profile IPOs, with companies looking to go public as...

TechCrunch Startups
Microsoft Research's Mirage gives video generation a persistent spatial memory that doesn't forget what's around the corner
Research & PapersMicrosoft

Microsoft Research's Mirage gives video generation a persistent spatial memory that doesn't forget what's around the corner

Microsoft Research's Mirage is a video world model that uses latent space storage instead of pixel-based point clouds...

The Decoder
Google Cloud's Open Knowledge Format turns scattered docs into Markdown files for AI agents
AI AgentsGoogle

Google Cloud's Open Knowledge Format turns scattered docs into Markdown files for AI agents

Google Cloud introduces Open Knowledge Format (OKF), a standardized specification that converts scattered...

The Decoder
KPMG fabricated AI case studies in a report designed to sell clients on AI adoption
Ethics & Regulation

KPMG fabricated AI case studies in a report designed to sell clients on AI adoption

KPMG published a report on AI adoption containing fabricated case studies involving UBS, the NHS, and other...

The Decoder
AI coding agents find the right file but miss the exact lines that matter, study shows
AI AgentsAnthropic

AI coding agents find the right file but miss the exact lines that matter, study shows

A new SWE-Explore benchmark reveals that AI coding agents like Claude and Codex can locate the correct files but...

The Decoder
Amazon and five other companies reportedly triggered the government crackdown on Anthropic's Fable model
Ethics & RegulationAnthropic

Amazon and five other companies reportedly triggered the government crackdown on Anthropic's Fable model

Amazon CEO and executives from five other tech companies reportedly warned the Trump administration about security...

The Decoder
Databricks Open-Sources Omnigent: A Meta-Harness That Composes, Governs, and Shares AI Agents Across Claude Code, Codex, and Pi
AI Agents

Databricks Open-Sources Omnigent: A Meta-Harness That Composes, Governs, and Shares AI Agents Across Claude Code, Codex, and Pi

Databricks has open-sourced Omnigent, a meta-harness platform that enables composition, governance, and sharing of AI...

MarkTechPost
As Anthropic suspends access to new models, India debates its AI future
Ethics & RegulationAnthropic

As Anthropic suspends access to new models, India debates its AI future

Anthropic has suspended access to new models, prompting tech leaders and policymakers in India to reassess the...

TechCrunch Startups
Amazon security research reportedly led to the White House’s Anthropic Fable ban
Ethics & RegulationAnthropic

Amazon security research reportedly led to the White House’s Anthropic Fable ban

Amazon's cybersecurity research reportedly identified vulnerabilities in Anthropic's Fable 5 AI model that could be...

The Verge AI
KPMG pulls report on AI usage due to apparent hallucinations
LLM

KPMG pulls report on AI usage due to apparent hallucinations

KPMG retracted a report on AI usage after discovering it contained hallucinations generated by AI tools. The incident...

TechCrunch AI
Amazon CEO reportedly raised Anthropic model concerns before government crackdown
Ethics & RegulationAnthropic

Amazon CEO reportedly raised Anthropic model concerns before government crackdown

Amazon CEO Andy Jassy reportedly raised security concerns about Anthropic's AI models that led the company to cut off...

TechCrunch AI
How to Build a QwenPaw Agent Workspace with Custom Skills, Model Providers, Console Access, and Streaming API Testing
AI Agents

How to Build a QwenPaw Agent Workspace with Custom Skills, Model Providers, Console Access, and Streaming API Testing

This tutorial guides users through building a QwenPaw agent workspace with custom skills, model provider integration,...

MarkTechPost
New AI model called "Count Anything" does exactly what it says, and that's harder than it sounds
Research & Papers

New AI model called "Count Anything" does exactly what it says, and that's harder than it sounds

"Count Anything" is a new AI model that can count objects in any type of image using text prompts, cutting error rates...

The Decoder