Insights archive
Every analysis we have published, newest first. The live feed covers the last 30 days; these pages go back to the beginning.
1302 stories · page 51 of 66
Google Cuts AI Subscription to $5, Undercutting Rivals
Google dropped its AI Plus subscription from $8 to $5 a month, doubling included storage, at the exact moment OpenAI and Anthropic are preparing for stock market listings where their pricing power will face public scrutiny for the first time.
· 2 min read
Anthropic Launches Claude Fable 5 With Hard Safety Limits
Anthropic has released Claude Fable 5, its most powerful AI model available to the public, but the model automatically blocks or redirects questions on cybersecurity, biology, and chemistry to an older, less capable model, a design choice that matters for any business exploring AI for technical or scientific work.
· 2 min read
Top Voice AI Tools Now Handle Bilingual Callers
A new benchmark reveals which AI voice tools can accurately transcribe customers who mix two languages mid-sentence, a common behavior among the world's 3.3 billion bilingual speakers that until now had no real industry test.
· 3 min read
Anthropic's Claude Opus 4.8 Now on Google Cloud
Anthropic's latest flagship AI model, Claude Opus 4.8, is now available on Google Cloud's enterprise platform alongside Claude Sonnet 4.6, giving businesses a new option for running complex, long-running document and workflow tasks inside infrastructure many already use.
· 2 min read
Google Launches Real-Time Voice Translator in 70 Languages
Google has released Gemini 3.5 Live Translate, a voice-to-voice translation tool that works across 70+ languages in near real-time, now available in the Google Translate app and coming to Google Meet for business users, which matters because it removes the need for human interpreters in everyday multilingual calls and meetings.
· 2 min read
Sandstone Raises $30M for In-House Legal AI
Sandstone, a startup building AI tools specifically for corporate legal departments at small and mid-sized businesses, has raised $30 million in Series A funding, entering a market where big AI players are now openly competing for the same lawyers.
· 2 min read
Google DeepMind Picks 15 European Robotics Startups
Google DeepMind has named the first cohort of its European robotics accelerator, a three-month, no-equity program giving 15 startups from nine countries access to Google's AI tools, cloud computing credits, and engineering expertise, at a moment when robotics investment in Europe more than doubled in 2025.
· 2 min read
Pentagon Labels Alibaba and Baidu as Military-Linked
The US Defense Department has formally designated Alibaba, Baidu, BYD, and dozens of other Chinese companies as linked to China's military, a move that will bar the Pentagon and its suppliers from doing business with them starting this month, and signals wider pressure on companies globally that still rely on these firms.
· 2 min read
AI Agents Are Joining Your Workforce, Ready or Not
Autonomous AI agents that can handle entire job tasks without human input are moving from pilot projects into real operations at large companies, and the organisations that figure out people management first will pull ahead of those still focused only on the technology.
· 3 min read
Doctors Can Be Sued for AI Mistakes Under Current UK Law
The Medical Protection Society is warning UK ministers that existing law holds doctors and hospitals legally responsible for harm caused by AI diagnostic tools, even when the error was the AI's, and that without reform this will slow adoption and expose clinicians to unfair liability.
· 2 min read
Harvard Study: AI Agents Cut Task Time by 87%
A joint Harvard and Perplexity study, using real data from over 84,000 sessions, finds that AI agents complete tasks 47 times faster than conversational AI search, at a fraction of the cost, and that users hand over more complex, higher-value work as they get comfortable with the technology.
· 3 min read
China Opens First Wind-Powered Underwater Data Center
China has put the world's first wind-powered underwater data center into full commercial operation off Shanghai, using seawater for cooling instead of electricity-hungry air conditioning, at a time when AI's appetite for power is making energy the single biggest constraint on how fast any country can build out AI infrastructure.
· 3 min read
UK Pilots AI Court Scheduler to Cut 80,000-Case Backlog
The UK government is trialling an AI scheduling assistant in crown courts to help clear a record backlog of 80,000 criminal cases, but lawyers warn the tool cannot substitute for money and staff the system has been starved of for over a decade.
· 3 min read
Apple Offers Free AI to Small App Developers
Apple launched a free AI tier for smaller app developers at its annual developer conference, removing the cloud usage costs that have been quietly derailing AI experimentation across the industry.
· 2 min read
Google NotebookLM Gets Gemini 3.5 and Code-Running Skills
Google has upgraded NotebookLM with its latest Gemini 3.5 model and built-in connections to Antigravity, its coding tool, so the research app can now write and run code inside your notebooks to do deeper analysis and produce richer outputs.
· 2 min read
AWS Launches Optimization Tools for Complex Operations
AWS published details on its mathematical optimization service, which helps companies find the best possible answer to complex operational decisions like routing, scheduling, and factory sequencing, with documented results including a 24% cost cut in logistics and a 10% gain in manufacturing output.
· 2 min read
Microsoft Launches Always-On Work Agent Scout for Microsoft 365
Microsoft has begun rolling out Scout, a background AI agent that acts on your behalf inside Teams, Outlook, and the rest of Microsoft 365 without waiting for you to ask it anything, marking a real shift from AI that answers questions to AI that takes action.
· 2 min read
Anthropic Plants Engineers Inside NSA for Cyberattacks
Anthropic has placed about six of its own engineers inside the U.S. National Security Agency to help deploy Mythos, a cyber AI model it refuses to release publicly, for offensive operations against foreign networks, even as it sues the Pentagon and files for a near-trillion-dollar IPO.
· 3 min read
NVIDIA Signs Major AI Factory Deals Across South Korea
During a high-profile visit to Seoul, NVIDIA CEO Jensen Huang signed deals with LG Group and Doosan Group to build AI-powered factories, train robots, and process data at industrial scale, extending a pattern of Korean conglomerate partnerships that now spans Samsung, SK, Hyundai, and more.
· 2 min read
UK Bets £500M to Keep AI Startups From Leaving
The UK government has launched a £500 million fund that combines cash, supercomputer access, and government contracts to stop homegrown AI companies from relocating to the US, while NVIDIA and CoreWeave pour billions into British data center infrastructure.
· 3 min read