TechDefused Newsroom AI
AI-assisted newsroom output reviewed by a named human editor before publication.
Articles
Instagram expands 13+ content rating and controls to India
Instagram is rolling its age-appropriate 13+ content rating and a new Limited Content setting out to Teen Accounts in India, automatically placing users under 18 into the updated setting and requiring parental permission to opt out.
Gemini adds notebooks to organize projects
Gemini introduces notebooks that let users collect chats, files and sources across apps, the company said in an announcement.
Apple's $599 MacBook Neo shocks PC market with budget pricing and A18 Pro chip
The most repairable MacBook in over a decade takes Apple into direct competition with Windows laptops and Chromebooks for the first time.
Michael Saylor says $60,000 Bitcoin bottom came from seller exhaustion
Michael Saylor attributes Bitcoin's $60,000 trough to "seller exhaustion" and "capital structure", News Explorer reported.
Amazon ends support for older Kindle devices
Longtime users posted frustrations on Reddit
AWS chief defends OpenAI investment as routine partner conflict after backing Anthropic
Matt Garman says Amazon's cloud unit has long-established processes for competing with the companies it also funds and supports
Anthropic launches Claude Managed Agents as enterprise revenue hits $30 billion run rate
The AI safety company is offering businesses ready-made cloud infrastructure for autonomous AI agents, with Notion, Rakuten and Asana among early adopters
Microsoft developer division head Julia Liuson to leave after 34 years
The executive, who oversaw the $7.5 billion GitHub acquisition, will step down in June and move to an advisory role
Claude suffers repeated outages as Anthropic battles Sonnet 4.6 errors across multiple days
Anthropic's AI chatbot has gone down several times in quick succession, with thousands of users reporting failures across chat, login and voice
Z.ai releases 754-billion-parameter open-source model built for hours-long autonomous work
GLM-5.1 is designed to sustain complex engineering tasks across thousands of tool calls without human intervention.
Memento-Skills lets AI agents rewrite their own capabilities without retraining underlying models
A university research framework uses a reflective learning loop to evolve agent skills dynamically, cutting data and maintenance costs
X rolls out Grok-powered translation and AI image editing tools
Elon Musk's social media platform is adding automatic post translation and a natural language photo editor to its app.
Google Colab adds Learn Mode and Custom Instructions
Google Colab is expanding its Gemini artificial intelligence integration with notebook-level Custom Instructions and a Learn Mode that guides users through coding.
YouTube bets on the living room as TV viewing tops 44% of watch time
The Google-owned platform is hiring engineers and designers to make television viewing more interactive across live and short-form formats
Alibaba opens chip-powered data centre as China pushes for AI self-sufficiency
The e-commerce giant and China Telecom have launched a facility in Guangdong running on domestically developed semiconductors
Meta launches Muse Spark as Zuckerberg bets on proprietary AI
The social media giant unveils its first major large language model since a $14 billion tie-up with Scale AI
Google Finance expands to 100+ countries
Google is rolling out an artificial intelligence-powered redesign of Google Finance with local-language support and new market tools.
British Bitcoin developer denies he is Satoshi Nakamoto after New York Times investigation
Adam Back says the newspaper's evidence amounts to coincidence. But the mystery of who created Bitcoin remains as unsolved as ever
Anthropic's most powerful AI model yet scans for security flaws, but could also create them
Mythos, a frontier model that Anthropic says exceeds anything it has previously built, makes its debut in a tightly controlled cybersecurity project. The timing is not accidental