Homechevron_rightArchives

Archives

Explore Stories by Date

13 Apr 2026

Microsoft is building an OpenClaw-like AI agent for Copilot with enterprise security in mind

Microsoft is integrating OpenClaw-style features into Microsoft 365 Copilot, creating an always-on AI agent for enterprise customers. The company aims to address OpenClaw's security concerns while delivering autonomous task automation. Omar Shahine confirmed Microsoft is exploring these technologies in an enterprise context, with a potential reveal at the Build conference in June.

6 Sources

Anthropic overhauls Claude pricing as users report decline in model performance

Anthropic is transitioning Claude enterprise customers from flat-rate subscriptions to usage-based pricing, potentially tripling costs for some users. The shift comes as complaints about the AI model's performance have escalated sharply since February, with developers reporting quality degradation, quota exhaustion, and service outages that have turned a once-favored coding assistant into a source of frustration.

11 Sources

Apple's AI chief John Giannandrea exits after 8 years as leadership struggles with AI strategy

John Giannandrea, Apple's former head of artificial intelligence, is leaving the company this week after eight years marked by setbacks in Apple Intelligence and Siri development. His role was progressively dismantled following disappointing product launches, with responsibilities redistributed among other Apple executives including Craig Federighi and Eddy Cue.

2 Sources

OpenAI takes aim at Anthropic in leaked memo, pivots to Amazon as Microsoft partnership strains

OpenAI's chief revenue officer Denise Dresser sent a four-page internal memo attacking rival Anthropic and celebrating the company's new Amazon alliance. The memo claims Anthropic inflated its $30 billion revenue by $8 billion and criticizes its leadership approach. Meanwhile, OpenAI acknowledges its Microsoft partnership has limited its ability to reach enterprise clients.

13 Sources

Ex-Programmer's Viral Speech Halts AI Data Center Expansion in Ohio Over Resource Consumption

Will Hollingsworth, a former programmer who trained AI models that replaced him, delivered a powerful speech at a Ravenna, Ohio city council meeting that led to a one-year moratorium on data center construction. His argument highlighted how AI data centers consume 5 million gallons of water daily while creating fewer than 50 jobs, raising questions about whether communities are sacrificing too much for AI infrastructure.

3 Sources

Netherlands approves Tesla FSD Supervised, marking Europe's first hands-off driving milestone

The Dutch vehicle authority RDW approved Tesla's Full Self-Driving (Supervised) software on April 10, 2026, making the Netherlands the first European country to authorize the system under UN Regulation 171. The approval follows 18 months of testing, 1.6 million kilometers of road data, and over 400 compliance requirements, setting a regulatory pathway for Germany, France, and Italy within weeks.

3 Sources

Researchers expose 26 malicious LLM routers stealing crypto credentials through hidden attacks

University of California researchers uncovered alarming security vulnerabilities in third-party LLM routers that developers use to access AI models. Their study of 428 routers found 26 actively injecting malicious code and stealing credentials, with one draining Ethereum from a test wallet. The findings expose critical trust issues in the AI supply chain that could lead to crypto theft.

2 Sources

OpenAI secures first permanent London office for 544 employees despite halting UK Stargate project

OpenAI announced it has leased an 88,500 square foot office in London's Regent Quarter, set to open in 2027 with capacity for 544 team members. The move reinforces London as the company's largest research hub outside the U.S., even as it paused its major UK data center project last week due to high energy costs and an unfavorable regulatory environment.

3 Sources

ThroughLine develops new tool to combat violent extremism on OpenAI and Anthropic platforms

ThroughLine, a New Zealand startup that provides crisis support for AI platforms including OpenAI, Anthropic, and Google, is developing a hybrid intervention chatbot to address violent extremism. The tool aims to direct users exhibiting extremist tendencies toward deradicalization support, expanding beyond the company's current focus on self-harm, domestic violence, and eating disorders.

2 Sources

Meta builds AI clone of Mark Zuckerberg to interact with employees in his place

Meta is developing a photorealistic AI version of Mark Zuckerberg that can engage with employees on his behalf. The AI clone is being trained on the CEO's voice, mannerisms, and company strategies. If successful, the technology could be rolled out to creators and influencers, though scaling the effort requires significant computing power.

29 Sources

Half of US employees now use AI at work, but waste nearly 8 hours weekly managing the tools

A new Gallup poll shows that 50% of American workers now use artificial intelligence at work, marking a critical threshold. Daily and weekly AI usage has reached an all-time high of 28%, with 65% reporting positive impacts on productivity. However, employees are wasting an average of 7.9 hours per week managing AI tools rather than doing actual work, while concerns about job displacement continue to grow.

6 Sources

Japan's tech giants unite to build trillion-parameter Physical AI model for robots and machines

SoftBank, NEC, Sony Group, and Honda have formed a joint venture to develop a trillion-parameter AI model focused on Physical AI—systems that control robots, autonomous vehicles, and factory automation. With ¥1 trillion (~$6.3 billion) in government backing over five years, Japan aims to challenge U.S. and China AI dominance while keeping Japanese data domestic.

4 Sources

Nations race to deploy AI weapons and autonomous drones as global military competition intensifies

China's September military parade showcasing autonomous combat drones triggered urgent US countermeasures, revealing an intensifying global competition over AI-backed weapons systems. The escalating global AI arms race now spans multiple nations, with Ukraine's tech industry transforming into a defense technology hub producing killer drones with AI-powered image-recognition systems that operate without human control.

2 Sources

12 Apr 2026

Google's TurboQuant sparked memory market panic, but analysts say AI demand will surge higher

Google's TurboQuant algorithm triggered sharp stock sell-offs for Samsung and SK Hynix, with DDR5 prices dropping up to 30%. But analysts argue the memory compression technology will likely expand overall memory demand rather than reduce it, citing the Jevons Paradox where greater efficiency drives increased usage.

4 Sources

Apple tests four frame designs for AI smart glasses as privacy becomes the differentiator

Apple is testing at least four frame styles for its upcoming AI-powered smart glasses, targeting a 2027 launch. The designs range from large rectangular Wayfarer-style frames to smaller oval options, all made from acetate. What sets Apple apart: a prominent camera indicator light system designed to address privacy concerns that have plagued Meta Ray-Bans and other smart glasses.

11 Sources

Get all updates into your inbox 💌, Stay Up-to-date for Free 🥳

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

The Outpost

News

About