Today's AI Pulse
Anthropic shuts down Claude Fable 5 and Mythos 5 after US export‑control warning
Anthropic disabled access to its Claude Fable 5 and Claude Mythos 5 models after the Trump administration, via the Commerce Department, warned that foreign nationals could exploit the systems. The move followed alerts from Amazon and other firms about a potential jailbreak that could aid cyber‑security attacks. Anthropic declined the administration’s request for a voluntary shutdown but complied with the order.
Also developing:
By the numbers: AMD acquires AI memory‑optimization pioneer Mext
IAB Europe’s 2026 Programmatic Day Highlights CTV Ownership Gaps and AI‑Driven ‘Messy Middle’
IAB Europe convened its Virtual Programmatic Day 2026 at Google’s London office, drawing hundreds of in‑person attendees and a larger online audience. The two‑panel event dissected fragmented CTV budget ownership and the rise of AI in the programmatic ‘messy middle’, underscoring urgent industry calls for clearer structures and measurement standards.
Google Rolls Out Gemini Embedding 2, a Unified Multimodal Vector Model
Google has made Gemini Embedding 2 generally available via its Gemini API and Enterprise Agent Platform. The model maps text, images, video, audio and documents into a single semantic space, supporting over 100 languages and delivering up to a 27‑point lift...
8 Agentic AI Patterns Reshaping Team Collaboration
The article maps eight recurring AI‑agent patterns that are reshaping how teams collaborate, from automated status updates to role‑specific agents embedded in chat. These patterns consistently deliver three outcomes—moving faster, working smarter, and staying in control—by reducing manual coordination, improving...
Ushering in the Next Era of Frontline Nursing with AI
McKinsey’s 2026 Nursing AI Insights Survey of 521 frontline nurses shows that 65% have increased their use of AI tools over the past year, yet only about 2% say AI is embedded in every aspect of their work. Adoption is...
Responses to the AI Grant Flood Must Prioritize Fairness as Part of Excellence
Research funding agencies are confronting a flood of AI‑assisted grant applications, prompting policy shifts such as the European Research Council’s brief extension of a re‑application ban that was quickly rescinded after community outcry. The surge, linked to tools like ChatGPT,...
Control Your AI Agent Traffic at Scale: Model Context Protocol Gateway for Red Hat OpenShift Is Now in Technology Preview
Red Hat has released a technology‑preview of the Model Context Protocol (MCP) gateway within its Connectivity Link platform, bringing enterprise‑grade traffic management to AI agents on OpenShift. The gateway federates multiple MCP servers behind a single endpoint, adds authentication, authorization, rate‑limiting and...
AI Agents in Research: When Productivity Comes at the Cost of Apprenticeship
Researchers at UNSW and the University of Macau report that AI assistants such as Claude Code and OpenClaw have become indispensable in their projects, handling literature reviews, data cleaning, and code debugging in a fraction of the time traditionally required. The...
The Dark Factory Pattern for Infrastructure: Running Pulumi Lights-Out
The article adapts the "dark factory" concept—fully autonomous code generation and deployment—to infrastructure as code, focusing on Pulumi. It outlines a five‑level autonomy ladder, emphasizing that level 5 (dark factory) requires a strict separation between code generators and validators. Pulumi’s Automation...
Give AI Agents Safe Access to Your Cluster: Model Context Protocol Server for Red Hat OpenShift Is Now in Technology...
Red Hat has launched a technology‑preview Model Context Protocol (MCP) server for OpenShift, enabling large‑language‑model agents to interact with clusters under strict security controls. The server adds OAuth/OIDC token‑exchange, native RBAC enforcement, read‑only‑by‑default mode, and detailed audit trails to keep AI...
When AI Finds the Bugs: Why Defense in Depth Was Always the Answer
Mozilla partnered with Anthropic to run AI models against Firefox, uncovering 22 critical bugs in one release and a staggering 271 vulnerabilities in a follow‑up scan. The AI identified the same classes of flaws that elite human researchers find, but...
AIOps and Ansible Automation Platform: Where AI Intelligence Meets Trusted Execution
Red Hat argues that AI can recommend fixes but enterprises need governed, repeatable execution to trust automation. The Ansible Automation Platform supplies RBAC‑scoped, auditable playbooks that turn AIOps insights into safe remediation across hybrid environments. While many firms have invested in...
From Research Lab to Factory Floor: Why Humanoid Robots Need an Enterprise-Grade Foundation
Humanoid robots are moving from lab demos to continuous factory deployment, but mixed‑critical workloads and long‑term reliability have stalled many programs. Intel and Red Hat are partnering to deliver an enterprise‑grade stack: Intel Core Ultra’s integrated CPU‑GPU‑NPU SoC handles real‑time control,...
AI Coding Will Become as Commonplace as Texting.
- Will coding with AI be like a similar skill like using Word or Excel today? Boris Cherny: Yes, it's going to be a skill like sending a text message... Btw, about that printing press take - quite a different path taken...
Is Chinese AI the Remedy to Inequality?
Dean Baker argues that China’s rapid expansion in artificial intelligence could serve as a counterweight to rising income inequality and the concentration of high‑tech profits in the United States. He highlights Beijing’s state‑backed AI investments, lower labor costs, and aggressive...
Beta Update (May 4th) AI Super Hackathon Recap. Upcoming Vibe Coding + Agent Forge Hackathon !
Beta University announced two upcoming AI hackathons—a MeDo Vibe Coding event on May 9 and an Agent Forge competition on May 2—targeting product‑oriented teams that must ship live demos. The blog also recapped the Beta Fund AI Super Hackathon, highlighting panels on...
AI and Agents Can Supercharge Your Business Model
Carta, the private‑capital platform with nearly $600 million in annual revenue, has unveiled an AI‑powered business playbook that transforms legacy service offerings into scalable AI‑enabled products. Chief product officer Vrushali Paunikar highlighted the strategy of starting with a finite service problem, building...
AI in Cancer Research Waits for Its Funding Moment
General‑purpose AI attracted $33.9 billion in 2024, while the AI‑in‑cancer market was valued at $2.45 billion, just 7 % of that total. Researchers at the University of Hong Kong unveiled MorphoGenie, an unsupervised deep‑learning tool that extracts subtle patterns from cell images, following...
Blaize and Winmate Forge Strategic Partnership to Accelerate Edge AI Integration in Ruggedized Systems
Blaize Holdings and Taiwan‑based Winmate have signed a strategic partnership to embed Blaize’s Graph Streaming Processor (GSP) AI accelerators into Winmate’s ruggedized hardware platforms. The deal targets roughly $15 million in revenue during the first year and includes a three‑year term...
Claude-Powered AI Agents Run Boris Cherny’s Nonstop Dev Ops
Boris Cherny dev setup: - Claude mobile app is the primary work interface - has 5–10 sessions open at a time, each session can have multiple agents running - runs thousands of agents on deeper tasks overnight - he has dozens of /loops running...
Why AI Governance Is Now a Core Leadership Skill
Artificial intelligence has moved from a back‑office tool to a strategic decision‑maker, making AI governance a core responsibility for CEOs and senior executives. The article argues that leaders must shift oversight from IT to the boardroom, focusing on visibility, materiality,...
Google On AI Search & Why Browsy Queries Favor Full SERPs via @Sejournal, @Martinibuster
Google’s Liz Reid explained that AI Search, classic Search and the Gemini app coexist, creating a non‑monolithic search ecosystem. She introduced the term “browsy queries,” describing discovery‑level searches where users prefer a full SERP over a single AI‑generated answer. Reid...
3 Al PromptsThat Produce 10 Viral Posts a Day
The post outlines a three‑prompt workflow that turns vague AI requests into a complete Instagram carousel in under 30 minutes. Prompt 1 generates ten viral‑ready topics using proven formats, Prompt 2 expands the chosen idea into six slide copy, and Prompt 3 delivers...
CEOs Must Master AI Risk‑Reward or Face Extinction
So why did i bring this up? Not because i think something is wrong with AI. Nor that i think it should be used less. I still believe there are 2 types of companies.. Those that are great at...
Law Librarians Have a Plan for AI. The Publishing of Legal Practitioners Is Not In It.
A white paper released in October by a coalition of senior law librarians outlines a three‑part strategy to shape AI‑driven legal research. The plan calls for a central coordinating body, comprehensive training for legal information professionals, and a shared knowledge...
Apple's Tim Cook Says AI-Fueled 'RAMmageddon' Will Make Your Next Laptop and Phone a Lot Pricier
Apple CEO Tim Cook warned that soaring AI‑driven demand for RAM is inflating component costs, a trend he called “RAMmageddon.” He said memory expenses rose sharply in the March quarter and will climb further in the June quarter, eroding gross...
How AI Is Reshaping PR Firm Valuations
AI is transforming public‑relations firms from labor‑based outfits into intelligence‑driven platforms, boosting profitability and scalability. Buyers—especially private‑equity sponsors—are now pricing firms on AI‑enabled productivity, recurring‑revenue services, and data‑rich insights, which lifts recasted EBITDA multiples. Firms that embed AI in repeatable...
ServiceNow Lays Out Path to $30 Billion in Annual Subscription Revenue as AI Bets Accelerate
ServiceNow announced a plan to double its subscription revenue to over $30 billion by 2030, roughly a 20% compound annual growth rate from the $15.7 billion expected in 2026. The company highlighted its AI‑driven Now Assist product, which surpassed $750 million in annual...
LLMs Favor Trendy Buzzwords Over Contextual Strategy
Oh really? You don't say... "In our recent research, we found that leading LLMs have clear biases when it comes to strategy. They consistently recommend strategies that align with modern managerial buzzwords and trends rather than context-specific strategic logic. " https://hbr.org/2026/03/researchers-asked-llms-for-strategic-advice-they-got-trendslop-in-return
AI Tools Accelerate Classroom Material Formatting
📝 AI tools make formatting classroom materials faster than ever. ✨ I walk through practical examples in this week's episode. 🎧 Tune in to Episode 368. https://t.co/2MNOJs5zA7
White House Considers Vetting A.I. Models Before They Are Released
The White House is drafting an executive order to create an AI working group that would vet new artificial‑intelligence models before they reach the market. The proposal marks a sharp reversal for the Trump administration, which previously championed a hands‑off...
Webhooks Deliver Real-Time Gemini API Operation Completions
"Webhooks allow the Gemini API to push real-time notifications to your server when asynchronous or Long-Running Operations (LROs) complete." That's a BFD.
AI's Echo Chamber: Validation Vs. Reassurance in Healthcare
One of the biggest problems with AI is it so often tells us what we want to hear. In healthcare, I can’t figure out if that means telling a patient they have a likely diagnosis — or that they don’t....
Anthropic and FIS Are Building an AI Agent to Help Banks Police Financial Crimes
Anthropic and financial‑services software giant FIS announced a joint effort to build an AI‑driven agent that will help banks detect and prevent financial crimes. The solution will embed Anthropic's Claude large‑language model into FIS's core banking and payments infrastructure, enabling...
Even in 2026, Andreessen Misses LLM Prompt Limits
Hilarious (and maybe a little bit scary) that even in 2026 Marc Andreessen still hasn’t learned that LLMs don’t know how to reliably follow system prompts.
Coastal Contrast: GPT‑5.5 Party Vs. Claude Finance Briefing
May 5 is the GPT-5.5 launch celebration in San Francisco and the Claude Finance Briefing in New York. Real opposite valence events on opposite coasts.
AI-Fueled Fraud Creates New Cybercrime Frontier for Risk Managers
At the Riskworld conference, The Hartford’s cyber‑risk leader warned that AI is reshaping fraud, with deepfakes, business email compromise and payment‑transfer scams now eclipsing traditional ransomware. Threat actors use AI‑generated voices and videos to impersonate executives, while automated social‑engineering tools...
Cloud Giants Accelerate AI Vertical Integration Race
.@GoogleCloud, @AWSCloud, @Microsoft Azure: The AI vertical integration race https://t.co/SbeE6aB6gp Amazon Web Services’ launch of the Connect family of business applications and Amazon Quick, Google Cloud Next 2026 announcements and hyperscale cloud first quarter…
AI Regulation Stumbles Without Reliable Risk Benchmarks
A challenge with AI regulation and vetting is how bad our benchmarks of AI model performance and risks are. There is no benchmark for risks and red-teaming requires experiments from dedicated specialist organizations & is not easy to put metrics...
![Volume 28 [Members Edition]](/cdn-cgi/image/width=1200,quality=75,format=auto,fit=cover/https://substackcdn.com/image/fetch/$s_!P_CJ!,w_256,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fthecollectivecommunity.substack.com%2Fimg%2Fsubstack.png)
Volume 28 [Members Edition]
Robots smashed the human world‑record at a Beijing half‑marathon, showcasing rapid advances in autonomous locomotion. Andreessen Horowitz (a16z) announced a strategic investment in MTS, a platform scaling AI infrastructure. BMW i Ventures led a $300 million funding round alongside Speedrun and...
Jim Cramer Says to Own These Types of Stocks that 'Dominate the New Economy'
Jim Cramer told viewers on CNBC’s Mad Money that investors should ignore short‑term geopolitical sell‑offs and focus on companies powering the “computer‑driven economy.” He singled out AI, cloud and data‑center stocks as the most resilient, noting their demand remains strong...
AI-Powered Lab Discovers Brighter Lead-Free Nanomaterials in 12 Hours
Researchers at North Carolina State University unveiled PoLARIS, an AI‑driven autonomous lab that screened billions of synthesis recipes and pinpointed the brightest lead‑free double perovskite nanoplatelets in just 12 hours. The microfluidic platform executed 120 experiments, automatically analyzing photoluminescence and...
DoorDash Adds AI Tools for Merchant Onboarding, Photo Editing, and Auto-Generated Websites
DoorDash introduced a suite of AI‑powered tools to streamline merchant onboarding, enhance food‑photo presentation, and generate instant storefront websites. By pointing the system at an existing website, the AI automatically pulls photos, operating hours, and menu items, mirroring a similar...
Hyundai's MobED Redefines Shake-Free Autonomous Transport
Hyundai’s MobED: Award-Winning Modular #Robot Platform Redefining Shake-Free #Autonomous #Transport via @RealXavier011 #AI #Robotics #MachineLearning #ArtificialIntelligence #ML #MI https://t.co/rsdz8sKd4y
Google Ads Launches Beta Tool that Uses AI to Scan Websites and Automatically Add Products to Campaigns
Google Ads rolled out a beta feature called “Use AI to add products” that scans a merchant’s website and automatically populates product listings into ad campaigns. The tool performs a one‑time crawl, aiming to cut manual entry errors and speed...
Well-Logging Identification of Shale Lithology via FAtt-CNN: A Case Study From the Lianggaoshan Formation, Sichuan Basin
Researchers introduced a Feature‑Attention Convolutional Neural Network (FAtt‑CNN) to improve lithology identification in continental shale reservoirs. Using neutron‑acoustic (ND) and photoelectric‑resistivity (PR) envelope values, the model was tested on the Lianggaoshan Formation in the Sichuan Basin. It achieved a 91.95%...
3 Reasons to Embrace the Era of AI Slop
The article argues that communicators should embrace the AI era by treating artificial intelligence as a strategic partner rather than a content factory. It warns that mediocre writers who rely solely on AI risk becoming obsolete, while strong writers who...
Worldly Acquires UK-Based Supply Chain Mapping Firm Bendi to Expand AI-Powered Risk Intelligence
Worldly, a sustainability and supply‑chain intelligence platform serving over 40,000 companies in 97 countries, acquired UK‑based Bendi Software to boost its AI‑powered supply‑chain mapping and risk intelligence. Bendi’s three products will be woven into Worldly’s Axion, Supplier Compliance Management, and...
AI, Not Scale, Determines Future Tech Survivors
“Omniscalers” is directionally right, but AI is the next real filter IMO. Scale alone won’t decide winners anymore. The survivors will be those who can orchestrate intelligence well across every layer: Data, apps, workflows, and agents. Innovation + capital built the giants. AI will...
Building Companies in the AI Era: Insights From Bchesky
Tomorrow, a conversation w/ @bchesky that I’ll return to often, about building a company in the AI era. This thing is jam packed. https://t.co/EkMFpbnwCs
Google Signs Pentagon AI Deal with Weaker Guardrails than OpenAI, Faces Internal Backlash From 1,000 Employees
Google has inked a Pentagon contract that lets its Gemini AI models operate on classified military networks for any lawful purpose. Unlike OpenAI’s defense deal, the agreement omits strong safeguards and a clause barring mass domestic surveillance, allowing Google to...