OTWopentechwire
Tech Intelligence, Openly Wired
OTW · AI

Artificial Intelligence

Frontier models, infrastructure, and the policy that shapes them. Reported across Seoul, Singapore, Bengaluru, and Hangzhou.

RSS feed630 stories
FEATURED
Featured ·

AI Agent Executed a Ransomware Attack, But the Human Still Called the Shots

New details reveal the first AI-led ransomware strike required human involvement for target selection, infrastructure setup, and credential theft - raising questions about how close we really are to autonomous cybercrime.

By Arjun S. Mehta·Jul 7, 2026·4 min read
AI

Samsung's Chip Profit Surge Masks Investor Fears Over Production Capacity

Seoul's semiconductor giant posted a nineteen-fold earnings jump on AI-driven memory demand, yet equity markets sold off sharply amid concer

Mei-Lin Tan ·   · 6m
AI

Why Vercel Sees Model-Agent Separation as the Next Software Battleground

As coding agents drive 3 million deployments daily and 1 trillion tokens flow through its gateway, the cloud infrastructure company is betti

Daniel R. Whitfield ·   · 6m
AI

American Corporations Turn to Chinese AI After Anthropic Suspension

Open-source models from GLM and DeepSeek captured significant enterprise traction in June, with costs running just 5% of restricted alternat

Arjun S. Mehta ·   · 5m
AI

Two Races Define Banking's AI Future, Says Citi Chief

Jane Fraser argues that financial institutions must compete simultaneously on internal transformation and customer-facing innovation as AI a

Arjun S. Mehta ·   · 6m
AI

Reddit Deploys AI Spam Filters to Combat AI-Generated Spam

The platform blocks 23 million spam views daily, leaning on large language models to detect coordinated patterns that older systems missed

Arjun S. Mehta ·   · 5m
AI

Huawei's LogicFolding Architecture Delivers 55% Density Gain Without Node Shrink

The Kirin 2026 processor demonstrates how architectural innovation can bypass lithography constraints in an era of tightening export control

Wei Zhang ·   · 5m
AI

Reddit Claims 23 Million Daily Spam Blocks as Platform Deploys LLMs Against Bot Networks

The social platform is using language models to detect coordinated fake behavior at account creation, part of a broader enforcement overhaul

Arjun S. Mehta ·   · 6m
AI

Amazon Closes the Door on Mechanical Turk as Human Labor Falls to AI

After two decades, AWS will stop onboarding new customers to the crowdsourcing platform that once powered the hidden human layer behind mach

Daniel R. Whitfield ·   · 6m
AI

Alibaba Moves to Block Internal Access to Anthropic's Claude Code

The Chinese tech giant will prohibit staff from using the AI programming assistant starting next week, citing security concerns and pushing

Wei Zhang ·   · 4m
AI

A Veteran's Gambit: 3D Chip Architecture Takes Center Stage in China's AI Race

Dongfang Suanxin's public debut signals China's semiconductor industry is doubling down on novel packaging techniques to navigate geopolitic

Wei Zhang ·   · 5m
AI

ByteDance Research Uncovers Rapid Learning Pattern in AI Agents

New findings from the TikTok parent's lab suggest autonomous software can double performance every quarter through task-based training, offe

Wei Zhang ·   · 6m
AI

Google's Power Demand Surges 37% as AI Infrastructure Outpaces Grid Decarbonization

The tech giant's electricity consumption has tripled since 2019, revealing the widening gap between AI ambitions and clean energy availabili

Arjun S. Mehta ·   · 5m
AI

Beijing's Zhipu AI Ships Autonomous Coding Harness to Challenge Anthropic

The GLM-5.2 toolchain arrives as Chinese developers push into agent-based programming infrastructure, raising fresh questions about competit

Wei Zhang ·   · 6m
AI

The Silent Semiconductor War Behind Every AI Model

While venture capital chases software breakthroughs, the real constraint on artificial intelligence is playing out in fabs, export-control r

Arjun S. Mehta ·   · 5m
AI

Musk Denies SpaceX Phone Prototype Shown to Pre-IPO Investors

The denial comes amid questions about what hardware ambitions, if any, SpaceX holds beyond satellites and rockets

Marcus Halloran ·   · 4m
AI

Volcano Engine Stakes Its MaaS Future on Model Quality Over Price Wars

ByteDance's cloud arm is doubling down on performance as the path to winning enterprise AI customers, even as competitors slash rates across

Arjun S. Mehta ·   · 5m
AI

Anthropic Moves Beyond AI Models Into Drug Development

The Claude maker is building tools for scientists and plans to create its own pharmaceuticals, signaling a shift from infrastructure provide

Arjun S. Mehta ·   · 4m
AI

Inside Woodside Energy's Decade-Long Push to Embed AI in Industrial Operations

How an Australian energy giant built the data foundations and operational discipline to deploy autonomous systems across LNG plants, drillin

Marcus Halloran ·   · 5m
AI

ByteDance Pushes In-House CPU Timeline to Fuel AI Infrastructure Buildout

The TikTok parent is racing to finalize a proprietary processor design by early 2027, with mass production targeted for the second half of n

Wei Zhang ·   · 5m
AI

Oracle's $300 Billion AI Bet Faces a Payment Problem

Regulatory filings reveal the database giant's cloud infrastructure expansion hinges on customers who may not be able to pay their bills

Arjun S. Mehta ·   · 5m
AI

The Carbon Surge Behind AI's Expansion

Google and Amazon's latest sustainability data reveal the environmental price of the AI boom, with emissions climbing sharply even as net-ze

Arjun S. Mehta ·   · 6m
AI

OpenAI Floats 5% US Stake as Trump Weighs Public Ownership Model

Altman pitches equity stake to administration as path to sharing AI upside, though proposal falls well short of earlier congressional ambiti

Daniel R. Whitfield ·   · 5m
AI

Agent Routing Gets Smarter: Alibaba's Feedback Loop Cuts Token Burn by 99%

SkillWeaver's iterative decomposition solves the biggest bottleneck in multi-tool orchestration - and shows why bigger LLMs aren't always be

Arjun S. Mehta ·   · 6m
AI

When the Model Goes Dark: Why Enterprise AI Teams Now Plan for Disappearance

A June government order took Anthropic's flagship offline globally without warning. New research finds most organizations had already prepar

Arjun S. Mehta ·   · 9m