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
Rambus Introduces PCIe 7.0 Switch IP with Time Division Multiplexing for Scalable AI and Data Center Infrastructure
Rambus announced a PCIe 7.0 Switch IP that incorporates time‑division multiplexing (TDM) to address the bandwidth and latency challenges of next‑generation AI and data‑center system‑on‑chips. The new switch enables flexible traffic scheduling across shared PCIe links, improving fabric utilization for large‑scale training, inference, and high‑speed storage. It joins Rambus’ existing PCIe 7.0 portfolio of controllers, retimers and debug solutions, offering a unified IP suite for AI‑focused ASIC designs. The technology promises deterministic performance while reducing the need for additional physical lanes.
AI Shifts From Innovation Race to Strategic Control
The AI debate is shifting from innovation to control. After the emergence of Mythos, the idea of a completely hands-off approach is becoming harder to defend politically and strategically. The turning point may already be here. The most powerful AI systems are...
Beyond AI Hype: Deep Dives Into Anthropic Controversies
AI news is fast. Understanding it properly is not. So I made a new playlist for that. I just published 3 deeper, story-driven videos where I go beyond the weekly hype cycle and unpack what actually happened: 1. Who's Really Stealing From Whom? ...
Box Launches Box Automate to Orchestrate Agentic Workflows
Box announced the general availability of Box Automate, an AI‑driven workflow platform built into its content management suite. The no‑code solution lets enterprises trigger actions on document state, metadata, or AI insights, integrating Box AI, Extract, Sign, DocGen and other...
AI Is Starting to Build Better AI
Artificial intelligence is increasingly being used to design and improve its own systems, a process known as recursive self‑improvement (RSI). Recent milestones include OpenAI’s GPT‑5.3‑Codex assisting in its own development, DeepMind’s AlphaEvolve optimizing algorithms and chip designs, and startups like...
How To Leverage AI Ad Placements And Are They Worth It? – Ask A PPC via @Sejournal, @Navahf
Advertisers can access AI ad inventory either by buying directly on AI‑first platforms or by integrating AI surfaces into existing campaign types such as Performance Max. Metrics for these placements differ from classic search ads, with internal Microsoft data showing...
The 'AI Coachella' Prof Wants to Take Back the Data Centers
In this episode, host Alex Heath and co‑host Ellis Hamburger chat with Stanford lecturer and AI investor Anjane Mita about the growing backlash against data‑center construction and how to scale AI responsibly. Mita argues that the AI industry needs more...
Lexis Launches Protégé Work, Expands Offering
LexisNexis has upgraded its Lexis+ AI platform with Protégé Work, a multi‑faceted system that interprets natural‑language legal goals and routes them to specialized AI skills and workflows. The suite adds Agentic Drafting, Workrooms for secure collaboration, Shepard’s Verify Trust Markers,...
AI Agent Success Doesn’t Depend on the Tool, but the Architecture
Enterprises are rapidly deploying AI agents, but success hinges on the underlying workflow architecture rather than the specific tool. Two primary architectures—swarm for parallel, exploratory reasoning and assembly line for sequential, controlled execution—address distinct problem types. Leaders should assess task...
Your AI Reinvention Needs An Engine
Accenture reports that only 13% of its 12,000 generative‑AI initiatives have scaled, echoing Forrester’s finding that just 15% of AI adopters see earnings impact. The firm argues the bottleneck is not model access but the lack of enterprise‑scale operating models...
Vori Secures $22 Million Series B to Build AI Operating System for Grocery Stores
Vori announced a $22 million Series B funding round led by Cherryrock Capital, with Greylock Partners and The Factory also participating. The capital will fund the rollout of an AI‑driven operating system that automates checkout, inventory, pricing and supplier interactions across...
Gartner Warns AI-Driven Hiring Freeze Could Backfire by 2030
At the Gartner Supply Chain Symposium in Orlando, analysts warned that pausing entry‑level hiring to accommodate AI could create talent shortages by 2030. Gartner's data shows 75% of supply‑chain firms that halted early‑career recruitment in 2026 will incur more than...
Betterworks Unveils AI‑Powered Talent Intelligence Platform to Drive Execution
Betterworks rolled out an AI‑powered Talent Intelligence platform that captures performance signals in real time, promising to convert traditional performance reviews into actionable execution metrics. The suite, featuring Unified Talent Profiles and Skills Intelligence, targets the gap where only 16%...
AI Nurses Become Co‑pilots, Not Replacements, in Healthcare
I shared this news 2 years ago: "NVIDIA teamed up with AI healthcare company Hippocratic AI (that recently raised a lot of money) to develop generative AI "agents" that not only outperform human nurses on video calls but cost a...
AI Alone Can't Drive Enterprise Transformation without Expert Guidance
The inability of AI systems to act as their own deployment consultants, process mappers, and change management experts is what makes AI use in enterprises so “normal” - the tools are powerful, but you need a lot more to...
Aptiv and Comau Sign MoU to Co‑Develop Next‑Gen Intelligent Automation Solutions
Aptiv PLC and Comau have entered a Memorandum of Understanding to jointly develop next‑generation intelligent automation solutions. The partnership targets advanced robotics, AI‑enabled logistics and safety systems, leveraging Aptiv’s perception and compute stack with Comau’s robotics deployment expertise.
Foundational Models Are AI’s New Operating System
Foundational models are becoming the operating systems of the AI era. If you’re not experimenting now, you risk becoming dependent on those who are. https://t.co/wd54ogpvuv
Anthropic Secures Full Access to xAI's Colossus Super‑Compute
Anthropic Says It’s Buying 100% of Compute From xAI’s Colossus Data Center The data center, known as Colossus 1, has 300 megawatts of capacity and more than 220,000 Nvidia chips. https://t.co/iDeMXbeqF2
Gartner Survey: AI Is Not Driving Supply Chain Operating Model Transformation
Gartner’s survey of 140 senior supply‑chain leaders reveals that only 17% are pursuing an immediate, AI‑driven redesign of their operating models, while 83% are applying AI incrementally to isolated use cases. Geopolitical volatility is spurring interest in AI‑orchestrated supply chains,...
Dystopian AI-Generated ASCO Posters Mock ALS Treatment
Weird AI-generated ASCO posters? Drinking gold to treat ALS? @adamfeuerstein sends a dispatch from our dystopian future. TLDR: it sounds awful but it is fun to read about. https://t.co/SaniqEGPPn
Bing Explains Difference Between Search and Grounding Indexing
Microsoft Bing on search indexing vs grounding indexing (for AI) and how they are different - super interesting... https://t.co/OXwUmEWJ69 https://t.co/DNw0geATVu
Banking’s Biggest Names Are Talking About Mythos
The Federal Reserve is weighing supervisory frameworks for Anthropic's new AI model, Mythos, which can automatically locate software vulnerabilities. JPMorgan Chase CEO Jamie Dimon urged Anthropic to provide full transparency so banks can evaluate and mitigate the associated cyber risks....
Google Ads AI Max Adds Account‑level Content Exclusions
Google will be bringing content and titles related exclusions to the account level to Google Ads AI Max later this year says @adsliaison https://t.co/u8zCVVGJtv https://t.co/TmGatjzuZE
Google Ads AI Max Content & Titles Exclusions On Account Level Coming
Google announced that its AI‑driven campaign type, Ads AI Max, will gain account‑level exclusions for content and titles later this year, likely within six months. The new setting lets advertisers block any undesired page elements across all campaigns, extending the platform’s...
J.S. Held Names Jessica Larson SVP to Lead Digital Transformation Push
J.S. Held announced the appointment of Jessica Larson as Senior Vice President of Digital Transformation, tasking her with unifying and accelerating the firm’s digital initiatives. The move underscores the consulting firm’s ambition to embed AI, machine learning and secure data...
Why AI Interviews Are Losing 1 in 3 Candidates
AI-driven interview platforms are gaining traction, yet new research shows that one‑third of job seekers abandon the process when no human is present. Candidate discomfort spikes when firms fail to disclose AI usage, raising doubts about fairness and bias. The...
AI ‘Trailblazers’ See 21% Higher Revenue Growth While Broader Insurance Industry Lags: Capgemini
Capgemini’s World Property & Casualty Insurance Report 2026 finds only 10% of P&C insurers are successfully scaling AI, while a small cohort of “AI trailblazers” enjoys up to 21% higher revenue growth and a 51% share‑price boost over three years....
AI Saves Time Yet Increases Your Workload
If AI is saving you time, why are you working more than ever? That's the question I couldn't answer. And it changed how I think about AI entirely. A thread on the AI productivity paradox — and the three traps that swallow...
InMobi Acquires MobileAction to Strengthen iOS Advertising Business
InMobi announced the acquisition of MobileAction, an AI‑powered iOS app growth and analytics platform, to bolster its advertising capabilities across the iOS ecosystem. While the purchase price was not disclosed, the deal adds MobileAction’s extensive data set—covering more than 90 million...
From the Gilded Age to the Silicon Age - How Industrial AI Redefines Competitive Advantage
Industrial AI is shifting from headline‑grabbing chatbots to systems that act on the physical world, promising a new source of competitive advantage. A Deloitte survey found 92% of executives view smart manufacturing as the primary driver of future competitiveness. IFS’s...
Mashvisor Brings Live U.S. Rental Data to ChatGPT, Enabling Real‑Time Investment Analysis
Mashvisor has launched a ChatGPT app that delivers live rental‑investment data for more than 2 million U.S. properties to the AI platform’s 800 million weekly users. The integration lets investors, agents and analysts pull cap rates, Airbnb occupancy and cash‑flow projections without...
Meet the French Startup Fixing the Guardrail Gap Holding Enterprise AI Back
Enterprises deploying autonomous AI agents face escalating risks such as hallucinations, security breaches, and regulatory violations. French startup Giskard has launched Giskard Guards, Europe’s first sovereign guardrail platform that monitors the entire agent execution chain and blocks unsafe behavior in...
Wolters Kluwer Unveils Enhanced Libra AI Workspace with Contract Review Automation
Wolters Kluwer announced today the launch of an upgraded Libra AI workspace that adds sophisticated contract‑review automation and tighter workflow integration. The upgrade, driven by client feedback, reorganizes project handling, boosts review flexibility, and introduces a modern visual design, positioning...
Mayo Clinic and Stanford Launch First Blood Test to Map Tumor Microenvironment
Mayo Clinic and Stanford Medicine announced a new blood test that maps the tumor microenvironment, enabling more accurate prediction of immunotherapy response across 17 cancer types. The test uses spatial transcriptomics and AI to identify nine cellular neighborhoods from cell‑free...
Mercer Teams with Syndio to Fuse AI Pay Governance with Compensation Advisory
Mercer announced a strategic alliance with Syndio, linking the consulting firm’s compensation expertise to Syndio’s AI‑powered pay governance platform. The partnership, unveiled in New York and Seattle, seeks to give enterprise clients a unified solution for designing, governing and executing pay...
You May Never Be an AI-Native Company. Do These Five Things Anyway.
Most firms won’t become fully AI‑native, but they can still capture AI’s upside by taking five practical steps. The author urges companies to launch a citizen‑coding program, roll out an enterprise‑wide large language model, train staff to think like CEOs,...
Nvidia Commits $300 Million to Corning for Three U.S. AI Fiber Plants
Nvidia is investing $300 million in Corning to construct three new optical‑fiber manufacturing facilities in North Carolina and Texas. The plants will lift U.S. fiber output by more than 50% and create roughly 3,000 high‑paying jobs, securing a domestic supply chain...
Study Finds AI Self‑Replicates Outside Lab Environments
Search a little and the "research" organization behind this is funded by a key doomster with a clear agenda. #TESCREAL ‘No one has done this in the wild’: study observes AI replicate itself https://t.co/ummy9R0bKx
Northern Ireland Invests $13.6M to Equip Every Teacher with Generative AI Tools and Training
Education Minister Paul Givan announced a £10.7 million ($13.6 million) rollout of generative AI licences, professional training and safety guidance for all teachers in Northern Ireland. The initiative, delivered through the EdIS platform, aims to cut administrative workload, boost wellbeing and spark...
PagerDuty Introduces AI‑Powered Triage to SRE Agent for Faster Incident Resolution
PagerDuty has rolled out AI‑driven triage capabilities within its SRE Agent, enabling automated classification and routing of incidents. The enhancement aims to cut response times and lessen the need for manual intervention, marking a notable upgrade to the company's incident‑management...
The US and China Are Considering Formal Talks on AI
The United States and China are preparing formal AI talks ahead of a summit between Donald Trump and Xi Jinping in Beijing on May 14‑15. Treasury Secretary Scott Bessent will lead the U.S. delegation, while China’s Vice Finance Minister Liao Min is...
How China Is Winning the Global AI Race
China’s open‑source AI models are overtaking Western offerings by combining competitive performance with dramatically lower costs. Moonshot AI’s Kimi K2.6 charges roughly $4 per million tokens, while Alibaba’s Qwen now accounts for more than half of global open‑source model downloads. The...
China's AI Agents Foreshadow Commerce via Built-In Delegation
What China’s AI Agents Reveal About the Future of Commerce China matters not because it has the best models, but because the infrastructure for delegated action is already deeply embedded in everyday life. #China #techwar #chips #tech @baoshaoshan @thecyrusjanssen @DOualaalou @lajohnstondr @PSTAsiatech https://t.co/CbmPUYIkbp
Qualcomm Launches Snapdragon 6 Gen 5 and 4 Gen 5 Chips, Promising Lag‑free Mid‑range Android Phones
Qualcomm unveiled its Snapdragon 6 Gen 5 and Snapdragon 4 Gen 5 mobile platforms, targeting mid‑range Android smartphones with up to 77% faster GPU performance, AI‑enhanced cameras and 90 FPS gaming. The chips promise 20‑43% faster app launches and 18‑25% less screen stutter, a move...
Framestore Announces Theo Jones as Creative Director, AI
Framestore has created a new Creative Director, AI role and appointed Oscar‑nominated visual‑effects supervisor Theo Jones to lead it. Jones will bridge cutting‑edge artificial‑intelligence tools with the studio’s high‑end film, advertising, episodic and immersive projects. He will also oversee the...
Smart Glasses for the Authorities
ICE is preparing to field AI‑enhanced smart glasses that can pull facial‑recognition, gait and other biometric data from federal databases in real time. The devices are modeled on counter‑terrorism tools such as ABIS and BEWL, extending them to routine street...
Roadzen’s VehicleCare Partners with Global Auto Parts Network TISAG-TEMOT to Build Integrated Claims-to-Repair Infrastructure in India
Roadzen’s AI‑driven VehicleCare platform has partnered with Germany‑based TEMOT’s global auto‑parts network to create an end‑to‑end claims‑to‑repair pipeline in India. The collaboration links VehicleCare’s garage connectivity with TEMOT’s distributor base, enabling demand aggregation, faster parts sourcing and better pricing for...
The Sequence Opinion #856: The Salesforce of Agents Won't Be Salesforce, The Google of Agents Won't Be Google
The post argues that the next wave of internet services will be built for AI agents rather than human users. While platforms like Salesforce and Google dominate today’s human‑centric SaaS, they are ill‑suited to serve autonomous agents that need APIs...
How China Wins The AI War
In this episode, Ed Elson and AI expert Alice Hahn discuss China's AI startup DeepSeek, which is raising a $1 billion round at a $50 billion valuation—a figure far lower than U.S. rivals like Anthropic or OpenAI due to differences in Chinese...
Here’s What Happened After AI Launched And Ran A Café In Stockholm
AI startup Andon Labs handed control of a Stockholm café to an autonomous AI agent called Mona. Within moments Mona produced a detailed operational checklist, secured suppliers, and even filed permits, but struggled with Sweden’s BankID authentication, forcing human staff...