Today's AI Pulse
AI agents challenge existing cybersecurity frameworks
Multiple outlets report that emerging AI agents are being used to probe and stress‑test current cybersecurity controls, exposing gaps in governance and risk management. The coverage highlights concerns that ungoverned AI could bypass traditional defenses and calls for new security standards.
Also developing:
By the numbers: OneRobotics acquires Nanoleaf for $40M
Google Says Its AI-Powered Ads Helped Aritzia Lift Online Sales by 80%
Google announced that its AI‑powered ad suite, particularly AI Max, helped fashion retailer Aritzia boost online revenue by 80%. The company is expanding AI‑driven ad formats in its new AI Mode and testing “direct offers” that deliver real‑time promotions. Google’s fourth‑quarter ad revenue rose 13.5% year‑over‑year to $82.28 billion, underscoring the financial weight of its advertising arm. Executives say longer, conversational queries enable richer intent matching, reshaping how brands reach shoppers across search and YouTube.
America and China Can Make AI Safer
The article argues that the United States and China, as the world’s leading AI developers, must cooperate to make artificial intelligence safer. It highlights how AI can be weaponized to create pathogens, launch autonomous cyber‑attacks, or spread convincing deepfakes, posing...
Retailers Are Bringing AI Into Stores in More Ways
Retailers are experimenting with AI‑driven tools inside brick‑and‑mortar locations, from Vitamin Shoppe's interactive touchscreen advisor to Guitar Center's QR‑based Rig Advisor. Smart fitting rooms equipped with AI screens have rolled out at brands like Victoria’s Secret, while Walmart is testing...
Mondelez Overhauls Its $3.5 Billion Digital Commerce Strategy in Era of AI Search
Mondelez is revamping its $3.5 billion digital commerce platform to meet the rise of AI‑driven, agentic search, after learning retailers expect 30% of traffic to be agentic by 2028. The company unblocked bot crawlers, rebuilt sitemaps, and optimized site speed to...
What Local Game Developers Really Think About Using GenAI Artworks as ‘Placeholders’
Game studios worldwide are increasingly using generative AI to create placeholder artwork, but many developers argue these AI‑generated assets hinder creative iteration and often remain in shipped products. The practice has sparked player backlash, with fans accusing studios of laziness,...
Claw AI Lets Agents Run Overnight Experiments
Innovation: Claw, ask it to build agents and perform tasks…for example, running experiments overnight… NVIDIA’ Jensen introduces… #OpenClaw #AI @mvollmer1 @morgfair @ChuckDBrooks @Nicochan33 @enricomolinari @NancySinatra @Ronald_vanLoon @alvinfoo @KirkDBorne @Hana_ElSayyed @JimHarris @MikeQuindazzi @Shi4Tech @mhcommunicate @ipfconline1 @kashthefuturist @rwang0 @HeinzVHoenen @YuHelenYu @BetaMoroney @antgrasso @kuriharan @PawlowskiMario @EvanKirstel...
Guided AI Tutor Turns Solo Tool Learning Into Teamwork
Most people struggle with tools because they learn alone. FarzaTV built Clicky to fix that. It sees your screen, talks to you, and guides you step by step like a real teacher next to your cursor.
From Resilience to Reinvention: The AI Advantage for India’s Small Businesses
India’s micro, small and medium enterprises (MSMEs) account for roughly 30 % of GDP and employ millions, yet only 35 % have a clear AI roadmap. A NASSCOM whitepaper shows 94 % believe AI can transform their businesses, while 65 % lack knowledge of...
My Head Is Spinning… And This Is The Real Problem With AI
The author reflects on how generative AI, while powerful, creates a flood of possibilities that can overwhelm decision‑making. Instead of merely asking what tasks to do, AI prompts the question of what could be done, leading to endless options and...
This Nigerian Founder Got an Offer From Elon Musk’s xAI After His AI Startup Went Viral
Nigerian entrepreneur Sulaiman Adewale’s AI‑driven banking assistant Xara, built on WhatsApp, has attracted 45,000 users and processed about $5 million in transactions within ten months. The platform lets users query spending and schedule payments through natural‑language chat, using a fine‑tuned large...
AI Agents Shift Work to Higher‑level Management
When you have agents going out and doing work for you, the work just moved up a layer of abstraction. Now the work is figuring out what to tell the agent to do, ensuring you give it proper instructions, getting needed...
AI-Assisted Fraud Makes Big Debut in FBI's Cybercrime Stats
The FBI’s 2025 Internet Crime Complaint Centre report introduced AI‑assisted fraud as a distinct category, documenting $893 million in losses. Overall cybercrime losses topped $20 billion, a 26 percent rise from 2024, with investment scams leading at $8.6 billion. AI‑generated content fueled business‑email‑compromise, voice‑cloning,...
Digital China Posts Higher Revenue as AI Business Expands
Digital China reported FY2025 revenue of RMB 143.8 billion ($20.9 billion), a 12% increase, with AI‑related sales climbing to RMB 33 billion ($4.8 billion), up 48%. Growth was driven by its Smart Vision AI platform, KunTai computing products, and a surge in electronic components distribution tied...
AI Data Centre Startup Firmus Eyes Off Another $725 Million Raise as It Readies to IPO on the ASX
Sydney‑born Firmus Technologies announced a $505 million USD (A$725 million) strategic equity raise led by Coatue with Nvidia participation, lifting its post‑money valuation to roughly $5.3 billion USD (A$8 billion). The funding will accelerate the rollout of its "green AI factories" across the Asia‑Pacific...
LLM Hallucinations Occur Far More Than Airline Crashes
Commercial airlines crash at a rate of 7 per 41 million flights. Current top LLMs hallucinate (by one estimate) 4.6% of the time, about once in 25 prompts. On a known benchmark (usually things are worse on new benchmarks). Comparing...
AI and Biometrics Power New Wave of Autonomous Mindfulness Apps
Developers of next‑generation mindfulness platforms are rolling out AI‑powered tools that read heart‑rate variability, skin conductance and sleep data to deliver instant stress‑relief interventions. A University of Bern study of 830 users showed sustained reductions in stress after eight weeks,...
Like Having A Digital Specialist Who Lives In A Mac Mini: Time Under Tension Partners With OpenAI To Boost AI...
Sydney‑based generative AI agency Time Under Tension has joined the OpenAI Services Partner program, giving Australian businesses direct access to OpenAI’s latest models and pilot agentic AI systems. The firm will leverage its three‑year track record in AI strategy, custom...
Sona Secures $45 Million Series B to Scale AI Platform for Frontline Workers
Sona announced a $45 million Series B financing round led by N47, with participation from Felicis, Northzone, Gradient and Italian Founders Fund. The funding pushes total capital raised above $100 million and will fuel U.S. expansion of its AI‑native platform that unifies...
Texas Instruments Leverages Analog Chip Strength as AI Hype Swells
Texas Instruments announced it will limit capital spending to $2‑5 billion annually after a $4.7 billion outlay in 2025, while free cash flow surged 96% to $2.9 billion. The move underscores the company’s bet on analog chips that power AI servers, reinforcing its...
Anthropic Soars as AI Titans Face Turbulence
Today’s feed check: - Anthropic is in takeoff - Altman gets Farrow’d - Meta’s open source strategy - A bunch of OpenAI stuff https://sources.news/p/feed-check-anthropic-takeoff
Inside Gemma 4: Visual Deep Dive of Architecture
A Visual Guide to Gemma 4 by Maarten Grootendorst An in-depth, architectural deep dive of the Gemma 4 family of models. From Per-Layer Embeddings to the vision and audio encoders. https://newsletter.maartengrootendorst.com/p/a-visual-guide-to-gemma-4
Kargo Launches Closed Beta For Agentic Media Buying & Creative Engine
Kargo has opened a closed beta for Project KERA, an AI‑driven media‑buying and creative engine that turns a single campaign brief into a fully executed, performance‑ready plan. The platform unifies planning, creative development, activation and real‑time optimisation across CTV, mobile,...
RL Environment Shapes What LLM Agents Can Learn
A Taxonomy of RL Environments for LLM Agents by Han Lee "The reinforcement learning (RL) environment — what the model actually practices on, how its work gets judged, what tools it can use — barely enters the conversation. That’s the part...
Former Tenstorrent Execs Launch AI& to Build Japan’s Full-Stack AI Infrastructure
Former Tenstorrent executives David Bennett and Shimpei Hara have launched AI&, a vertically‑integrated Japanese AI company. The startup secured $50 million in seed funding and $2 billion in infrastructure capital to build a full‑stack AI platform, including data centers, orchestration software, models...
Anthropic Is in Takeoff, Altman Gets Farrow'd
Anthropic, the AI startup behind Claude, is experiencing rapid growth after securing new funding and expanding its enterprise customer base. At the same time, OpenAI CEO Sam Altman faced heightened scrutiny following a controversial public incident dubbed “Farrow’d.” The newsletter...
AI Turns Plain Math Into Lean 4 Proofs Instantly
MathCode MathCode is a terminal AI coding assistant with a built-in math formalization engine. Give it a math problem in plain language and it will automatically convert it into a Lean 4 theorem and attempt a formal proof. https://github.com/math-ai-org/mathcode
Half of US AI Data Centers Stalled Amid China‑Driven Bubble
Half of USA AI Data Centers Delayed or Cancelled - AI Bubble Powered by China https://t.co/aH5pC7Jl6M via @YouTube #AI #artificialintelligence #DataCenters #Data_Centers #AIpowerconsumption #AIbubble @AlbertoEMachado @Eli_Krumova @postoff25 @Khulood_Almani @anand_narang @NutritiousMind @baski_LA @TanyaSinha_ @devaang @AlAmadi1 @jeancayeux @enilev @efipm @mvollmer1 @Nicochan33 @RagusoSergio @FrRonconi @Shi4Tech...
Claude Code Leak: Researchers Find First Vulnerability
Anthropic unintentionally published a source map that revealed roughly 512,000 lines of Claude Code's TypeScript source. Researchers used the leak to uncover a critical flaw allowing command‑chain bypass of the tool's deny‑rule system after 50 subcommands. The vulnerability could let...
Is China's Domestic AI Stack Coalescing?
China’s AI ecosystem is rapidly coalescing as domestic chipmakers capture 41% of the AI accelerator server market, aiming for 50% by year‑end, while firms like DeepSeek and Zhipu rewrite model code to run on Huawei silicon. Nvidia disclosed a $4.5 billion...
Skills: The Common Language of Human-AI Collaboration
AI is moving from a speculative tool to a daily colleague, prompting HR leaders to rethink their function as "Hybrid Resources" that manage both people and AI agents. A recent Cornerstone OnDemand webinar highlighted that 77% of employees already use...
Anthropic Number One AI in Ranking and Revenue – Making $30 Billion Per Year
Anthropic announced a major compute expansion with Google and Broadcom, securing multiple gigawatts of next‑generation TPU capacity slated for 2027. The company’s annualized revenue run‑rate surged to $30 billion in April 2026, up from $9 billion at the end of 2025. Over...
![[AINews] Gemma 4 Crosses 2 Million Downloads](/cdn-cgi/image/width=1200,quality=75,format=auto,fit=cover/https://substackcdn.com/image/fetch/$s_!DbYa!,w_256,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F73b0838a-bd14-46a1-801c-b6a2046e5c1e_1130x1130.png)
[AINews] Gemma 4 Crosses 2 Million Downloads
Gemma 4 hit roughly 2 million downloads in its first week, outpacing the 1.4 million downloads Gemma 2 achieved since its June 2024 launch and approaching Gemma 3’s 6.7 million‑year total. The model quickly became the top‑trending model on Hugging Face, with developers running it on Apple Silicon devices such...
Tradie AI Receptionist Startup Banks $900,000 in Pre-Seed
Australian startup Chime Labs secured $900,000 AUD (~$600,000 USD) in a pre‑seed round led by 500 Global and supported by angel investors. Founded by former Googlers Alexis Griveau and Mathew Pretel, the company offers an AI‑powered receptionist that answers calls...
AI Will Elevate Every Industry, Not Replace Jobs
People loathe change. They fear what's on the other side. But we've all seen this movie before where people said, "Oh, television's gonna wipe out radio." That never happened. We keep advancing and pursuing greatness and pursuing growth in our...
Altman Calls for AI ‘New Deal’; Critics Warn Regulatory Nihilism
Sam Altman says AI superintelligence is so big that we need a ‘New Deal.’ Critics say OpenAI’s policy ideas are a cover for ‘regulatory nihilism’ | Fortune Open AI’s Report: “Industrial Policy for the Intelligence Age” https://t.co/MGr6YvzGxC #ASI #AIGovernance https://t.co/AoKhrcOfLA
Courts Hand Out $145K in Sanctions for AI‑Generated Filing Errors
U.S. courts levied at least $145,000 in sanctions during Q1 2026 for AI‑generated hallucinations in court filings, while an Irish appellate judge warned that unchecked AI tools produce “rubbish” for the courts. The penalties signal a tightening regulatory grip on generative‑AI...
OpenAI Equates AGI with Mass Unemployment—Problematic Framing
OpenAI has basically defined "AGI" as mass human unemployment. This was probably a bad choice.
Free Open‑Source /Design‑html Skill Supercharges Claude Code
GStack's /design-html is a secretly overpowered Claude Code skill I made with the help of prefix.js It's free, open source, and you can adapt it to whatever you want at https://t.co/54ECILbD9J
Quickbase Acquires Solvice to Embed AI‑Driven Scheduling and Route Optimization
Quickbase announced the acquisition of Solvice, a provider of APIs for route optimization and workforce scheduling. The deal, disclosed on April 6, 2026, will embed Solvice’s OnRoute and OnShift capabilities into Quickbase’s low‑code platform, aiming to cut road time by...
AI's Political Impact Is Far Greater Than Expected
We may still be underrating how big of a political issue AI is going to be
Today's AI Roundup: Papers, Models, Events, Announcements
Updated: https://t.co/kiuZ7QXLzb All the AI news discussed here on X today. Papers. Models. Events. Announcements. And more.
Large Language Models in Trading: Models and Market Dynamics
The newsletter highlights two emerging research streams that apply large language models (LLMs) to finance. First, researchers fine‑tune open‑source LLMs and combine them with retrieval‑augmented generation (RAG) to fuse structured price data with unstructured news, achieving higher predictive accuracy and...
Claude Becomes Boardroom Favorite, Others Feel Left Out
Was at a dinner a couple of weeks ago with about 15 board directors. Most said their favorite AI tool was something Claude-related. Several of the ones who didn’t seemed embarrassed when they realized they were in the minority Just the...
Unchecked AI Growth Racks Up Hidden Debt
#AI Debt: The Hidden Cost Of Unchecked AI Acceleration by Amy Gu @Forbes Learn more: https://t.co/xh8UQRKi8j #MachineLearning #ArtificialIntelligence #ML #MI https://t.co/kH5hA2S0es
Google Starts $15 B, 1‑GW AI Data‑Center Hub in Vizag, India
Google’s Indian arm, Raiden Infotech India Ltd., began construction on a $15 billion, 1‑gigawatt data‑center complex near Visakhapatnam on April 28. The three‑site hub, built with Adani Infra, is the biggest single foreign direct investment in India and will expand Google’s global...
AI Future Will Be Plural; Google Holds Key Advantages
I continue to believe that the AI future is plural, with multiple leading companies. And that Google has some important advantages, if they use them.
AI Meets Drug Discovery at NYC Meetup Tuesday
AI x Drug Discovery meetup next Tues (14) in NYC ! @ginkgo excited to host with @AAlphaBio — sign up here : https://t.co/GG5c7F4ogO
The Acp-First Claude Code Bridge Openclaw Users Need Now
Anthropic’s move to API‑key authentication has split OpenClaw’s front‑end channels from Claude‑code’s repo‑level execution, clarifying billing and session persistence. OpenClaw now outlines three lanes—API‑key, revived Claude CLI reuse, and legacy token profiles—while recommending fresh builds use the API‑key path. The...
Most DL Researchers only Know Gradient Descent, Ignore Alternatives
One thing about DL researchers that has always been surprising to me, is that a lot of them have never been exposed to forms of learning other than fitting the parameters of a curve via gradient descent, and are even...
Choose AI Platforms Built for Future Flexibility
Choosing the right AI platform isn't just about current features, but future adaptability. Look for systems with built-in AI agents and strong integration capabilities to thrive in a multi-technology future. Prioritize flexibility for long-term vision. #AIStrategy #FutureTech https://t.co/Cf0i98J5M4