Today's AI Pulse
Malaysia launches $440M sovereign AI cloud to boost security
Malaysia is committing roughly $440 million to build a sovereign AI cloud, aiming to keep sensitive data such as military and intelligence information within the country while still welcoming global cloud providers like Google, Microsoft and AWS.
Also developing:
By the numbers: ASML invests $1.4B in French AI startup Mistral AI
Microsoft Introduces 'Ask Copilot' Box to Windows 11's Taskbar – but only if You Want AI There
Microsoft announced an opt‑in "Ask Copilot" box for the Windows 11 taskbar in the latest preview build (26220.7051) for Beta and Dev channels. The feature replaces the traditional search box with an AI‑powered input that can surface files, apps, and suggest queries, while still offering a click‑through to classic Windows Search. It also adds one‑click icons for Copilot Vision and voice commands, and is disabled by default, requiring users to enable it via Settings. The rollout is limited to the preview and uses existing Windows search APIs, so it does not expand Copilot’s access to personal data.
LLMs Show a “Highly Unreliable” Capacity to Describe Their Own Internal Processes
Anthropic’s new study on “Emergent Introspective Awareness in Large Language Models” finds that current LLMs are largely unreliable at describing their own internal processes, with the best‑performing models (Opus 4 and Opus 4.1) correctly identifying injected concepts only 20‑42% of the time....
Younger Workers Are Skipping Meetings - and Trusting AI to Take Notes for Them
Younger workers, particularly Gen Z, are turning to AI note‑taking tools instead of attending meetings, with 19% of all workers and 26% of hybrid employees using them frequently, compared with 13% of in‑person staff. The Software Finder study links frequent...
Some AI Browsers Can Bypass Publisher Paywalls, Report Says
AI web browsers such as OpenAI’s ChatGPT Atlas and Perplexity’s Comet can circumvent publisher paywalls, successfully retrieving a 9,000‑word subscriber‑only feature from MIT Technology Review that standard ChatGPT could not access. By masquerading as ordinary users, these agents evade robots‑exclusion...
AI Email Client Auto‑Tags, Scores, and Drafts Replies
this weekend, i vibe coded a prototype email client that i want... - for each email, looks up contacts details + notes in attio, then org details, looks at lists to determine if portfolio, LP, etc. - summarizes this context and creates...
OpenAI Wants Your Next Security Researcher to Be a Bot - New Aardvark Tool Finds and Fixes Software Flaws Automatically
OpenAI has launched Aardvark, an autonomous AI agent built on ChatGPT that scans code, runs tests and proposes patches to fix software vulnerabilities at scale. In private‑beta testing the tool achieved a 92% success rate on benchmark “golden” repositories, and...
AI's Learning Speed Will Eclipse Human Progress Soon
Here I have a cute argument with Grok about whether it is a world model. As a kid a neighbor and I built our own sailboat. Then raced it on the San Francisco Bay. In three years @Tesla_Optimus will build a faster...
Speed Over Discernment Turns AI Into Empty Waste
Today’s AI slop was yesterday’s food. Before algorithms ate our words, farmers fed leftovers to pigs. It was a closed loop. What people couldn’t use became nourishment for something else. Nothing was wasted because nothing could be. Then industry found shortcuts. Distillers...
Studio Ghibli, Bandai Namco, Square Enix Demand OpenAI Stop Using Their Content to Train AI
The Content Overseas Distribution Association (CODA), representing Japanese IP owners such as Studio Ghibli, Bandai Namco and Square Enix, sent a formal letter to OpenAI demanding that the company stop using their copyrighted works to train its text‑to‑video model Sora 2....
Coca-Cola’s New AI Holiday Ad Is a Sloppy Eyesore
Coca‑Cola rolled out a new AI‑generated holiday commercial titled “Holidays Are Coming,” featuring a mishmash of animated critters that critics say look sloppy and inconsistent, though the iconic Coke truck wheels finally turn. The ad was produced with AI studios...
Big Tech's AI Twins: Your Identity Becomes Their Profit
Meta, Google, Apple - they're all building AI replicas that capture your face, expressions, movements, personality. This goes way beyond Face ID. They're basically creating a version of you that knows you better than you know yourself. The fidelity is remarkable...
OpenAI Signs Massive AI Compute Deal with Amazon
OpenAI has signed a seven‑year, $38 billion agreement with Amazon Web Services to supply cloud compute for ChatGPT, Sora and future models. The deal grants access to hundreds of thousands of Nvidia GPUs, with capacity slated to be online by the...
The AI Industry Is Running on FOMO
Big‑tech firms Amazon, Google, Microsoft and Meta are collectively spending more than $350 billion on AI this year and expect the figure to top $400 billion next year, while dedicated AI players such as OpenAI are generating $12 billion in revenue but face...
Dia’s AI Browser Starts Adding Arc’s ‘Greatest Hits’ to Its Feature Set
Dia, the AI‑native web browser acquired by Atlassian for $610 million, will incorporate several of Arc’s most popular features such as sidebar mode, vertical tabs, pinned‑grid view and Google Meet picture‑in‑picture, while leveraging a new architecture that promises better speed, security...
Sell Remote‑Operated Home Helpers, Not False Robot Dreams
Companies selling the dream of autonomous household humanoid robots today would be better off embracing reality and selling “remote operated household help”. Have teams of employees running them 24/7, with the option to reduce their workload as autonomous behaviors become...
I Put Napster View AI on My MacBook Pro, and I'm Now Convinced No One Needs This Much Face Time
Napster View is a $99 hardware add‑on for M1‑based MacBooks that projects a circular holographic screen above the laptop, housing AI‑driven avatars that act as a “digital team” of specialists and can even create a personal digital twin. The system integrates...
AI Coaches Will Transform Preventative Health in Decade
AI coaching is going to revolutionize preventative health Whoop now tracks bloodwork and biomarkers, combining with sleep, strain, and recovery data for personalized AI plans It's just the start, but in 10 years everyone will probably have a personal AI coach/doctor that's...
Adobe Warns Workforces Risk Being Left Behind without Creative and AI Skills
Adobe’s latest research shows 19% of UK workers feel they lack the creative and digital skills needed for an AI‑driven future, while 65% of job seekers consider AI expertise essential for landing their ideal role. A third of hiring managers...
Tesla's AI Paves Way for Autonomous Humanoid Revolution
It won’t matter. Now that I see what Tesla and xAI is up to, building a world model, that knows EVERYTHING, I am 100% convinced that Tesla will be the world leader in humanoids. And will be all autonomous within three years. I...
Top Marketing Legends Share AI Lead‑Gen Secrets – Free Webinar
Imagine an event where your heroes of digital marketing are all presenting. That event is happening on Wednesday. My heroes will be there. It's a free webinar totally focused on lead gen and conversion. I'm doing a session on AI...
The State of AI: Is China About to Win the Race?
The Financial Times and MIT Technology Review debate the AI rivalry between the United States and China, noting that China now leads in AI citations (22.6% of global citations) and patents (69.7% of filings) and has overtaken the U.S. in...
Buy AI Unless Your Data Is a True Moat
When should you buy AI vs build AI? Most teams start with the wrong assumption: that building means innovation. In reality, it usually means distraction. Buying wins when speed, scale, and integration matter. Building only pays off when your data...
AI Belongs to Humanity, Not Superpowers
Professor Himanshu Tyagi critiques Donald Trump’s AI Action Plan, arguing that deregulation and a nationalist competition mindset misinterpret how AI advances. He stresses that AI’s progress depends on global talent, open‑source collaboration and diverse data, citing that immigrants founded 55 % of...
OpenAI Strikes $38 Billion AI Training Deal with Amazon
OpenAI signed a seven‑year, $38 billion agreement with Amazon Web Services that grants the AI firm access to hundreds of thousands of Nvidia GPUs, with all capacity slated for deployment by the end of 2026 and the option to expand through...
Five Ways to Build a Data Foundation that Actually Lasts
The article outlines five essential strategies for creating a durable data foundation that powers AI, beginning with unifying customer identities across systems to eliminate fragmented profiles. It stresses the need for clean, up‑to‑date data to prevent AI failures, recommends a...
The Rise of Agentic AI in Cybersecurity
Agentic AI, autonomous systems that can reason, adapt and act without constant prompting, is poised to become a major disruptor in cybersecurity, with forecasts that one‑third of enterprise applications will embed such agents by 2028. The technology promises to alleviate...
Windows 11 Is Finally Getting a Handy Bluetooth Headphone Sharing Feature for some – Here's How It Works
Microsoft has rolled out a preview of a new "shared audio" feature in Windows 11 (build 26220.7051) that lets two Bluetooth LE Audio headsets receive the same sound stream from a single PC. The capability is accessed via a Quick Settings tile and...
Google Has a New Experimental DeepMind AI Tool to Help Your SMB Scale Marketing Fast
Google Labs, in partnership with DeepMind, has launched Pomelli, an experimental AI marketing platform aimed at small and medium‑size businesses. The tool builds a brand profile from a company’s website, then suggests custom campaign ideas and generates on‑brand assets that...
Turn ChatGPT Into a Friendly Therapist with Simple Prompts
3 AI Prompts That Turn ChatGPT Into Your Friend and Therapist Exploring unconventional uses of AI—this article shares practical prompts to shift ChatGPT from a tool to a companion or support aid, with caution around boundaries and ethics. Read more...
Microsoft Inks $9.7B Deal with Australia’s IREN for AI Cloud Capacity
Microsoft signed a five‑year, $9.7 billion contract with Australia’s IREN to secure AI cloud capacity built on Nvidia’s GB300 GPUs, to be deployed at IREN’s 750‑megawatt facility in Childress, Texas, through 2026. IREN is concurrently investing about $5.8 billion in GPUs and...
LG Founder’s Grandson, Production Firm Partner up to Bring AI to Filmmaking
LG Group founder’s grandson Brian Koo, via investment firm Stock Farm Road, has launched a 50‑50 joint venture with AI film producer Utopai Studios called Utopai East. The JV will develop AI‑focused infrastructure, anchored by a planned 3‑GW data center...
BMW Gets Green Light for ‘Hands-Off’ Highway Driving in Major First – Here’s How It Works
BMW became the first automaker to obtain approval under the new UN Regulation No. 171 for Driver Control Assistance Systems, enabling its Level 2 Motorway Assistant to be offered in a broader range of markets. The system, featured on the upcoming iX3...
Google Pulls AI Model After Senator Says It Fabricated Assault Allegation
Google removed its Gemma AI model from the AI Studio platform after Republican Senator Marsha Blackburn alleged the model fabricated false criminal accusations about her. Gemma, designed for developers and accessible via API, was being used by non‑developers to answer...
AI Reasoning: From Pattern Prediction to Human‑like Insight
🧩 Can AI really think – or is it just predicting patterns? In this video, I dive into one of the most fascinating areas of artificial intelligence: AI reasoning. From logical problem-solving and pattern recognition to understanding cause and effect,...
Amazon Partnership Fuels More NVIDIA Chips for OpenAI Scaling
Very pleased to be working with Amazon to bring a lot more NVIDIA chips online for OpenAI to keep scaling!
Meet Denario, the AI ‘Research Assistant’ that Is Already Getting Its Own Papers Published
An international research team released Denario, an open‑source AI system that autonomously conducts end‑to‑end scientific research, generating publishable papers in about 30 minutes for roughly $4 each. The modular architecture uses specialized agents for idea generation, literature review, methodology design,...
OpenAI Commits $38B with AWS for Faster ChatGPT
[🚨 BREAKING] OpenAI bets big on AWS. A $38B partnership to make ChatGPT faster, smarter, and ready to scale. → EC2 UltraServers with NVIDIA GB200s for large-scale training capabilities → Tens of millions of CPUs for real-time speed → Built for the next generation...
AI Drafts Legal Docs, Lawyers Still Essential
ai can draft legal docs but you still want a lawyer to review it. i’m a fan (and investor) of services like @DocDraftai that combines the best of both into a single elegant service :)
Google's World Model Paradoxically Benefits Apple
Google is building a world model that knows everything too. This is a big win for Apple, even though it looks like a loss.
Adapt to AI or Risk Becoming Redundant
Will AI replace you, or empower you? 🤔 Artificial Intelligence: Change, Adapt and Evolve or Become Redundant https://t.co/MXS9sqG0f4 A thought-provoking article, by @grattonboy #AI #FutureOfWork @grattongirl https://t.co/vhroANardT
Open-Source AI Guide: 200+ Pages, Full Transparency
No excuse anymore not to train your own models! This 200+ pages with full transparency. Let's go open-source AI!
Gemma AI Uncovers Novel Cancer Therapy Pathway
How a Gemma model helped discover a new potential cancer therapy pathway https://t.co/lMiONmkR5m @GoogleDeepMind #ai #iot #5g #agenticai #ces2026 #Healthcare https://t.co/jGcrTLlnfO
EU AI Rhetoric vs Reality: Dignity Claims Questioned
why do European politicians always say things like “the unique aspect of ai built in Europe is that it respects human dignity and values” as opposed to American ai that doesn’t? do you even use ai? even if that were true...
5G, IoT, and Agentic AI Enable Self‑Thinking Networks
Connectivity is no longer about linking devices, it’s about linking intelligence. When 5G, IoT, and Agentic AI converge, networks start thinking for themselves. 👉 https://t.co/95n8oVHSQl #AI #5G #IoT #CES2026 https://t.co/kiZ8vpLpqa
OpenAI Unveils ChatGPT Safeguards with Parental Controls
OpenAI details new ChatGPT safeguards including parental controls https://t.co/EIialPYVrz @perplexity_ai #AI #IoT #ChatGPT https://t.co/KcgcMvYKqf
AI Leads Top Downloaded Business Education Papers
In an analysis produced by SSRN for the Financial Times, Artificial Intelligence dominates the top 10 most downloaded business education papers. https://t.co/8ZpSWt0PEM #SSRN #Research #AI #Elsevier #Researchers #TechNews
Boost Student AI Literacy with Library Guide
Are you looking to elevate your students' AI literacy skills? Explore our AI literacy guide to help you incorporate AI into your library’s literacy programs and services. https://t.co/cpNzKcYQDa #ScopusAI #GenAI #ResponsibleAI #Elsevier
PharmaPendium AI Delivers Next‑gen Regulatory Intelligence
PharmaPendium AI is changing the game in regulatory intelligence. Built by Elsevier, it combines deep domain expertise with powerful AI to deliver. https://t.co/7V4QOlk0JI #PharmaAI #Pharma #DrugDevelopment #AIinPharma #Elsevier #LifeSciences #Innovation
Swiss LLM First EU AI Act‑Compliant, Despite Non‑EU Status
european ai innovation…the swiss make the first llm compliant with the eu ai act, while not being in the eu themselves. https://t.co/1ybB1jcNcG
Generative AI Investment Surges to $33.9B, Global Adoption Rises
🧠 Stanford’s 2025 AI Index😀 ➡️ Generative‑AI investment reached $33.9B in 2024 (up 18.7%) ➡️ 78% of organizations use AI ➡️ The U.S. still leads, but China is catching up #AIIndex #AITrends #GenAI https://t.co/8IK0WMDuiA https://t.co/pFmN6NiYUH