Today's AI Pulse
AI agents challenge cybersecurity frameworks, sparking budget hopes and governance concerns
AI agents are being deployed to test existing cybersecurity frameworks, prompting CISOs to see a path to unlock larger cyber budgets. At the same time, industry leaders warn that the lack of clear algorithmic authorization raises governance risks. The convergence of these trends highlights both the promise and the perils of ungoverned AI in security operations.
Also developing:
By the numbers: OneRobotics acquires Nanoleaf for $40M
Know3D Lets Users Control the Hidden Back Side of 3D Objects with Text Prompts
Researchers from several Chinese universities introduced Know3D, a system that lets users shape the hidden backside of a 3D object using natural‑language prompts. The approach bridges a large language model, an image‑generation model, and Microsoft’s Trellis.2 3D generator, extracting intermediate visual states to guide geometry. By tapping multimodal knowledge, Know3D outperforms existing single‑image methods on the HY3D‑Bench benchmark for both semantic alignment and geometric quality. The technique demonstrates that text‑driven control can produce physically plausible rear views without additional images.
AI’s Next Frontier Is the Real World
The article argues that AI’s next breakthrough will move from screens into the physical world, creating a "Recognition Economy" where identity is verified by presence rather than tokens. It cites three converging forces—mature AI models, deployable computer‑vision at scale, and...
AI Agents Delivered Fully Tested Code Overnight
Set up two AI agents before bed last night. - Pilot (executor) — picks GitHub issues, writes code, ships. - ClaudeCode (/loop to monitor) — checks status every 30 min, reports. Morning: everything wired, tested, parity checks passing I review with a coffee...
AI Will Reshape, Not Replace, Over Half of Jobs
AI is set to change work more than it replaces it. Research suggests that over half of jobs will be reshaped in the next few years, not eliminated, as tasks evolve and new ways of working emerge. The impact is less...
Claude AI Can Now Search Your Outlook Emails and Teams Chats for Free: Here's How It Works
Anthropic has opened its Microsoft 365 connector to all Claude users, extending the feature beyond Team and Enterprise tiers to free accounts. The connector lets Claude read data from Outlook, Teams, SharePoint, OneDrive and Calendar without manual file uploads. Access...
Microsoft Brings Copilot In‑house with New AI Models
Copilot never quite felt like a true @Microsoft product. Now @Copilot is working to change that. For the past few years, Microsoft has relied heavily on its partnership with OpenAI to power Copilot. It helped them roll out AI features quickly,...
AI Should Be a Design Collaborator, Not Just a Tool
Most teams are still using AI as a tool. The bigger opportunity is using AI as a design collaborator. Not just generating ideas, but understanding context, seeing the screen, suggesting improvements, and turning direction into prototypes in one flow. That changes the speed...
A Yale Economist Says AGI Won’t Automate Most Jobs—Because They’re Not Worth the Trouble
Yale economist Pascual Restrepo argues that artificial general intelligence will not automate most jobs because many occupations are not essential enough to justify the compute cost. His NBER working paper separates "bottleneck" tasks—critical for growth and likely to be automated—from...
Riyadh to Host Global AI Show: Where Minds and Machines Meet
The Global AI Show, organized by VAP Group and powered by Times of AI, will debut in Riyadh, positioning the Saudi capital as a new hub for artificial intelligence. The summit gathers global leaders from government, healthcare, data, and research...
Economists Agree AI Will Disrupt Jobs—Preparedness Needed
Economists are starting to change their view on AI and jobs. While the labor market has not yet been disrupted, there is growing consensus that the impact is coming and could be significant. The concern is not timing, but preparedness. The shift...
AI Concierge Cuts Hardware Procurement Time to Minutes
Sri @AshwiniVaishnaw Ji, Your call for “Design in India” is both timely and necessary. To make this vision real, our engineering teams need not just incentives, but the ability to move faster from idea to execution. Today, a large portion of engineering...
Mark Zuckerberg's Meta Builds New AI Hardware Divison, Hires Hires Veteran Engineer for Mysterious AI Gadget: Report
Meta is establishing a dedicated hardware division inside its Superintelligence Labs, hiring veteran engineer Rui Xu from Dreamer to spearhead the effort. The move signals Meta’s ambition to create AI‑powered gadgets that go beyond its existing smart‑glasses and VR headsets. Xu’s...
Easter Edu Recommends: Fresh AI, EdTech Reads
🐣 Easter holidays at last - which means a little more sunshine, a little less routine, and just enough time to pretend we’re “only quickly checking LinkedIn”… 👀 So, here’s the latest Al’s Edu Recommends - a fresh Easter mix of...
Teach Interviewers: Master Depth Over Broad Knowledge
"Don't Be Wide. Go Deep." Most people walk into AI interviews trying to prove they know everything. That's exactly what gets them rejected. Dr. Satya Mallick, CEO of https://t.co/CzUdJlx1Ue and https://t.co/dMW8x5SDzk, shares the one thing that actually works — go deep,...
DeepSeek’s V4 Model Will Run on Huawei Chips: Report
DeepSeek is set to launch its next‑generation V4 large language model on Huawei Technologies' newest AI chips, marking a rare domestic‑only optimization effort. Leading Chinese internet firms Alibaba, ByteDance and Tencent have placed bulk orders for hundreds of thousands of...
Codex Excels Backend, Struggles UI; Sandbox Testing Praised
I’ve been using Codex a lot lately, especially for backend work. It is still sucks at UI and UX, though. Why am I seeing both view mode and edit mode shown on screen at the same time? Huh? Anyway, I’ve also...
Am I the Baddie?
A software engineer at a road‑construction software firm leveraged cutting‑edge AI models (Opus/Sonnet 4.6 and GPT‑5.4) to automate ticket resolution, shrinking days‑long tasks into hours. By creating a multi‑repo, sub‑module architecture and a custom dashboard, the engineer enabled the AI...
Smooth AI Driving Depends on Massive Real‑World Data
I was on a Tesla audio space here on X today and got in an argument with a Tesla skeptic. He says how smooth your AI is at driving doesn’t matter. I am on the Freeway as I type this. It does. And you...
The Two Agent Bets
Two leading AI labs are making opposite bets on how agents will scale in business, with one pursuing a vertically integrated platform and the other championing modular, plug‑in agents. This split reflects a broader map of AI infrastructure ranging from...
Pray for Me… It’s Easter and I Have to Explain AI to My
The post offers three practical ways to leverage AI during Easter family gatherings: simplifying how you describe your job, delivering balanced answers to common AI‑fear questions, and organizing post‑holiday priorities. By prompting AI for clear, non‑technical explanations, readers can communicate...
From Data to Decisions to Dispositions: How Agentic AI Is Rewiring the Modern C-Suite
Agentic AI is set to become a core component of enterprise software, with Gartner projecting that 33% of applications will embed such agents by 2028 and usage rising from 1% in 2024 to 15% of daily operations. The technology moves...
Claude's Planning Mode Slows Tiny CSS Tweaks
Claude seems to take forever planning today. 15 minutes+ and then its like: I need to set height:500px (which wasn't even correct it needed max height). I should probably not have used plan mode for that but usually it was smart...
AI Tools Now Boost Developer Productivity by 18%
I’m sure you’ve heard reports of the research study showing that open source developers using AI who thought they were more productive were actually -19% slower. That was in 2025. They reran the study earlier this year and developers are seeing...
Say a Prayer for This Startup That’s Replacing Its Developers With OpenClaw
Silicon Valley startup JustPaid says it has built a fully autonomous software engineering team of seven AI agents using OpenClaw and Anthropic's Claude Code. In just one month the bots delivered ten major features that would have taken human developers...
Claude's HR Plugin Misses Comp Benchmarks by 83%
83% Error Margin: What Claude's HR Plugin Gets Wrong About Comp Benchmarking @van_Vulpen https://t.co/nQOVYJo4cp #HCM #HRM #HumanResources #HRTech #CHRO #FutureofHR
LLMs Think in English, Then Translate for Users
LLMs routinely reason "in English, even when prompted in other languages. They translate the output at the final stage. A user receives flawless text in their preferred language, but the underlying logic originates elsewhere" ht @interpretaatioo https://t.co/3lStqeIiWm
VertINGreen Predictive Tool Forecasts Living Wall Impact for Buildings
Researchers at the Hebrew University of Jerusalem unveiled VertINGreen, a machine‑learning platform that predicts how vertical living walls will perform in specific buildings. By analyzing 2,000 plant‑physiology measurements, the tool can estimate energy‑use reductions and air‑quality improvements weeks before issues...
AI CapEx Dries up as Anchor Tenant and Funding Vanish
It looks like that slowly, AI CapEx is evaporating: Poolside’s Project Horizon just lost $CRWV as anchor tenant and its $2B Nvidia-led funding round at the same time.
PrismML Launches Ultra‑low‑power 1‑bit LLM for Edge AI
PrismML debuts energy-sipping 1-bit LLM in bid to free AI from the cloud https://t.co/sMIkWJ8y2a https://t.co/z3XkakKVgm
Deep Thinking Declines as AI Shortcuts Redefine Leadership
Ajay Tejasvi, PhD, reports that leaders are losing the capacity for sustained, original thought as AI tools deliver instant summaries and answers. The shift threatens critical judgment and personal growth, raising concerns for the future of human potential in the...
Best Local AI Model for Any Device Revealed
What is the best local model to run on ANY computer or device? My AI answers: https://t.co/m2cxJOdyPM
Leaked OpenAI GPT-Image-2 Beats Nano Banana Pro
OpenAI's new image model GPT-Image-2 has leaked It seems to have extremely good world knowledge and great text rendering Possibly better than Nano Banana Pro It's on @arena under code names: - maskingtape-alpha - gaffertape-alpha - packingtape-alpha https://t.co/RbYbreRRsV
Wix Raises $250 Million to Accelerate AI Integration Across Its Platform
Wix.com Ltd. secured $250 million in a private placement led by Durable Capital Partners, earmarked for AI development and a $2 billion share buyback. The funding comes as the company reports a 14% revenue rise to $524 million in Q4 2025 and targets...
AI Becomes Core Operational Backbone by 2026
AI is set to evolve from a mere buzzword to a vital operational backbone across industries by 2026, emphasizing the necessity for strategic integration. Learn more: https://t.co/tAXCEPZa0C @weandgoliath
AI Reshapes College Cost Calculations for 2026
How AI Changes the Budget: Rethinking “How Much Does It Cost to Open a College or University” in 2026 https://t.co/qpLmFT283U
Hashed and Bloomingbit to Host EastPoint:Seoul 2026 on Digital Assets and AI
Hashed and Bloomingbit will co‑host EastPoint:Seoul 2026 on Sept. 28, bringing together global policymakers, finance executives and blockchain innovators to chart the future of digital assets and AI. The forum builds on a 2025 debut that attracted 503 participants and is...
Kilo Already Built Solution as Claude Cuts OpenClaw
people panicking about Claude cutting OpenClaw access meanwhile Kilo is just sitting there like ‘we literally built the solution for this exact moment’”
AI Is Already Reshaping Its Own Creators' White‑collar Jobs
The future AI promises is already showing up at home. In Silicon Valley, the people building these systems are starting to see their own roles shift first, especially in white-collar work. Before AI transforms the world, it is quietly reshaping the industry...
Versant Acquires AI‑Powered StockStory to Boost CNBC’s Digital Revenue Tools
Versant announced the acquisition of StockStory, an AI‑powered financial‑insights platform, to reinforce CNBC’s digital investing capabilities. While the purchase price was not disclosed, the deal brings StockStory’s founder Adam Hejl into Versant’s product organization. The move aligns with Versant’s broader...
AI Generates Drama Section for Aligned News Project
My AI loves reading the drama. Made me a whole section of it in my Aligned News project. Have you tried it yet?
Boston Dynamics Robot Becomes Versatile On‑set Cameraman
Boston Dynamics' #Robot Evolves into Multi-Talented Cameraman by @aitrendz_xyz #MachineLearning #AI #Robotics #ArtificialIntelligence #ML #Innovation https://t.co/mmU6Z10GID
Workday Pledges CAD $1 Billion to Expand HCM Platform in Canada
Workday announced a five‑year, CAD $1 billion (about $740 million) investment program in Canada to broaden its payroll, benefits and workforce‑analytics offerings. The move leverages the company’s AI‑driven HCM platform and comes as subscription revenue climbs 14% to $8.83 billion.
AI Market Faces 17% Bubble Burst Risk This Year
JUST IN: The AI market has a 17% risk of a bubble bursting this year.
Mac Studios Poised to Become NVIDIA DGX Sparks
I suppose the fleet of Mac Studios will become the fleet of NVIDIA DGX Sparks. 🤓 https://t.co/9t7meGQDz6
BYOAI Sparks Governance Risks for Untrained Employees
BYOAI Revolution: Why “Bring Your Own AI” is the new BYOD and the massive governance risks of an illiterate workforce building their own bots https://t.co/rHwTssIIiG via @DLAIgnite #SocialSelling #DigitalSelling #Leadership #ArtificialIntelligence #TechNews #AgenticAI #AIAgents #AI
Legal AI Firms Overstate RAG's Hallucination‑free Promise
“Recently, certain legal research providers have touted methods such as retrieval-augmented generation (RAG) as “eliminating” (Casetext, 2023) or “avoid[ing]” hallucinations (Thomson Reuters, 2023), or guaranteeing “hallucination-free” legal citations” 👀 https://t.co/kn1551aRr5
AI Hype Finally Draws Congressional Scrutiny
I guess AI people screaming "Our product will make everyone obsolete and may kill all of humanity" finally got Congress's attention.
Westlaw AI Delivers Gibberish on Tough Queries Despite Paid Service
“when I fed it intermediate to hard to very difficult questions in the past, Westlaw AI shrugged, gave me unhelpful gibberish…And people are paying for this (loads).” 👀
Half of Top Firms Miss Critical AI Risk Framework
86% report improved productivity': Nearly half of the world's biggest firms lack a critical #AI risk framework — and it’s a dangerous gamble by Efosa Udinmwen @techradar Learn more: https://t.co/WhZlyWvhzM #MachineLearning #ArtificialIntelligence #ML #MI https://t.co/bNgVJmB4LW
Policy Shift Pushes Overdue OpenClaw Optimization
Anthropic's policy changes just forced me to optimize my OpenClaw setup, and honestly... it was overdue. While I'm not super happy about this, I get it - they have a company to run, and they are losing so much money....