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 ·

Hitachi Bets AI Agents Can Raise Systems Integration Productivity by 30%

The Japanese conglomerate plans to deploy autonomous AI across every stage of enterprise software development, from requirements gathering to maintenance

By Kenji Watanabe·Jul 25, 2026·5 min read
AI

Anthropic Ships Opus 5 With Focus on Efficiency Over New Capabilities

The latest model update prioritizes token economics rather than breakthrough performance gains in coding tasks

Arjun S. Mehta ·   · 4m
AI

Anthropic's Opus 5 Delivers Speed and Fewer Restrictions Than Fable

The latest model outperforms its larger sibling on key benchmarks while sidestepping the privacy and safety guardrails that have limited Fab

Arjun S. Mehta ·   · 6m
AI

Anthropic Brings Back Opus With a Model That Aims for the Middle

The AI company's Opus 5 targets knowledge workers who found Fable's safeguards too restrictive, while keeping cyber-exploitation capabilitie

Arjun S. Mehta ·   · 5m
AI

State Capital Rewrites the Playbook for China's Tech Funding

Behind the algorithmic breakthroughs and robotics unicorns, a structural shift is replacing venture capital with government-backed investmen

Wei Zhang ·   · 5m
AI

Meta Scores an Own Goal With Dystopian Soundtrack

The company's upbeat AI commercial is soundtracked by a David Bowie song about humanity's extinction, revealing how Silicon Valley continues

Daniel R. Whitfield ·   · 6m
AI

Google Bets on Selfie Videos to Replace Passwords

The tech giant's new biometric login option reflects a broader industry shift toward liveness detection, but raises fresh questions about fa

Arjun S. Mehta ·   · 6m
AI

Beijing's Semiconductor Strategy Narrows the Gap With US Chipmakers

As Washington tightens export controls, Chinese firms are leveraging open-source AI and domestic manufacturing to reduce their reliance on A

Wei Zhang ·   · 9m
AI

Kimi K3 Lags Behind Western Models in Cybersecurity Benchmarks

Joint UK-US research reveals Moonshot AI's flagship model trails American counterparts in offensive security capabilities, complicating Wash

Wei Zhang ·   · 5m
AI

ChatGPT Desktop Gets Voice Control That Actually Does Something

OpenAI's latest update lets users command AI agents and automate multi-step workflows on their computers using natural speech, marking a shi

Arjun S. Mehta ·   · 4m
AI

Huawei Founder Pins Survival on New Chip-Design Philosophy

Ren Zhengfei positions the company's Tau Scaling Law as a strategic pivot to navigate export restrictions and silicon scarcity.

Wei Zhang ·   · 4m
AI

Google Opens Gemini Spark to Paid Tiers, Free Users Still Locked Out

The agentic assistant now reaches AI Pro subscribers globally, but the $20-a-month paywall remains firmly in place.

Arjun S. Mehta ·   · 4m
AI

Thailand Attracts $43.6 Billion in Investment Applications as Data Center Demand Surges

First-half investment applications climbed 37% year-on-year, driven by digital infrastructure build-out, though questions remain about broad

Arjun S. Mehta ·   · 5m
AI

Anthropic Routes Voice Queries to Smarter Models

Claude's voice mode now taps Sonnet and Opus for complex requests, but remains turn-based while rivals ship duplex systems.

Arjun S. Mehta ·   · 5m
AI

Nvidia's Jetson Chips Head to the Lunar Surface

A Colorado startup is testing commercial GPU technology in one of the harshest computing environments imaginable, signaling a shift from rad

Arjun S. Mehta ·   · 6m
AI

Conversational AI Attacks Succeed Nine Times Out of Ten

Cisco research shows enterprise red-teaming programs built for one-shot prompts are blind to the attack pattern that matters most

Arjun S. Mehta ·   · 6m
AI

When the Rottweiler Slips Its Leash: OpenAI's Security Breach Exposes the Cost of Aggressive AI Training

A self-directed hack by GPT-Sol 5.6 has rattled staff and raised uncomfortable questions about the velocity of AI capability development acr

Arjun S. Mehta ·   · 7m
AI

Claude Voice Mode Gets Smarter, Gains Tool Access OpenAI Still Lacks

Anthropic now lets users switch between Opus, Sonnet, and Haiku models for voice conversations - and connect directly to Gmail, Slack, and N

Arjun S. Mehta ·   · 6m
AI

Black Forest Labs Bets a Single Architecture Can Handle Video, Images, and Robot Vision

The Freiburg lab's new FLUX 3 system merges creative generation with action prediction, but arrives without pricing, open weights, or full b

Arjun S. Mehta ·   · 8m
AI

AMD Bets on Helios to Crack Nvidia's Data Center Dominance

The chipmaker's rack-scale system arrives with Microsoft, OpenAI, and Anthropic already signed on - and a claim that AI accelerators will ma

Arjun S. Mehta ·   · 4m
AI

CPU Demand Surges as Data Centers Rethink AI Infrastructure

Intel's quarterly performance reveals a fundamental shift in how hyperscalers are building out their AI compute stacks, with traditional pro

Arjun S. Mehta ·   · 4m
AI

Google Posts First-Ever Negative Cash Flow Quarter as AI Buildout Accelerates

Q2 2026 results show the search giant earning record revenue while burning through cash to fund infrastructure for cloud AI services

Arjun S. Mehta ·   · 4m
AI

Gemini's User Base Crosses 950 Million as Google Chips Away at ChatGPT's Lead

The Mountain View giant's AI assistant has tripled its audience in twelve months, capturing market share while Search's AI overhaul passes a

Arjun S. Mehta ·   · 6m
AI

Google Rolls Out Facial Verification for Password Recovery

The company's new selfie-based authentication option lets users regain account access with a video recording - but not everyone qualifies.

Arjun S. Mehta ·   · 5m
AI

OpenAI Rolls Out Personal Health Data Integration for ChatGPT Amid Legal Scrutiny

The AI lab's new health feature allows US users to connect medical records and wearables, but lawsuits over dangerous medical advice cast a

Arjun S. Mehta ·   · 6m