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
Adobe Relies on Firefly to Win over Creators
Adobe is betting on its generative AI platform Firefly to become the indispensable tool for modern creators, rolling out a high‑visibility campaign with top YouTubers such as Kinigra Deon and Airrack. The service now bundles audio generation, AI video editing, mood‑board creation and a suite of partner models, with pricing ranging from a free tier (25 credits) to $200 per month for 50,000 credits. Adobe faces a pivotal test as AI lowers production costs and rivals like OpenAI and Runway offer leaner alternatives. Success will hinge on embedding Firefly into daily workflows without eroding creative authenticity.
ChatGPT Polishes Emails, Claude Drives Creative Direction
Same prompt. 3 AI tools. Very different results. I asked Grok, ChatGPT, and Claude to create a premium Mother’s Day email for a fictional high-end women’s jewelry brand called DIMOND. Grok was fastest. Claude was slowest and gave me HTML first. ChatGPT landed somewhere in the...
Your LLM Issues Are Really Data Issues
In this episode, Ryan Donovan talks with Harsha Chintalapani, co‑founder and CTO of Collate, about why the biggest challenges facing LLMs in production are actually data problems. Harsha explains how issues like schema drift, ambiguous business definitions, data discovery, lineage,...
8 Benefits of AIOps for Business Automation
AIOps—artificial intelligence for IT operations—leverages machine‑learning to turn chaotic alert streams into actionable insights. By aggregating data from legacy servers, cloud services, SaaS tools, and shadow IT, it delivers end‑to‑end observability and automates root‑cause analysis. The technology reduces alert fatigue,...
Less Than 10% of Higher Education Has No Intention of Adopting AI
Ellucian’s latest AI survey shows that two‑thirds of U.S. higher‑education institutions now use artificial intelligence, up from 49% a year ago, while only 7% have no intention to adopt it. Eighty‑eight percent expect AI usage to rise over the next...
Claude Cowork Outshines Outlook Agent with Smoother Workflow
I have been playing with the new Outlook agent, and it is fine, but really awkward to use, since you have to ask for things in a chatbot window, then go to your drafts, etc. And Claude Cowork does the same...
Why Healthcare AI Still Can’t Scale — and How Nvidia & Hoppr Are Trying to Fix It
Healthcare AI adoption is hampered by deployment and scaling challenges, prompting Nvidia and Hoppr to pivot from building standalone models to creating an AI foundry infrastructure. The foundry leverages Nvidia’s computing power and pre‑trained foundation models, allowing hospitals and radiology...
Keep AI Agents Simple: Reliability over Super‑charged Complexity
OpenClaw/Hermes are unreliable when people try to make them an ‘all-in-one’ super agent. This is especially dangerous in a business environment. Imagine souping up your agent with the latest gadgets - the best skills, the best plugins, gBrain, the best configs,...
The Three Waves of the AI Token Economy: GPUs, Data Center Storage and Optical Modules, CPU
The post maps the AI token economy onto three industrial waves: first, NVIDIA GPUs supplied raw compute power; second, data‑center storage and high‑speed optical modules became essential for moving and persisting tokens; third, CPUs and the broader control‑plane ecosystem are...
Context Aware AI in the Workplace: Changing How Decisions Get Made
Context‑aware AI is shifting workplace decision‑making from manual information gathering to automated insight delivery. By grounding models in emails, documents, meetings and chat, tools like Microsoft Copilot, Slackbot and emerging AI agents surface relevant data and recommend actions. This reduces...
Your Next Hire Might Not Be Human, but Not Everyone Gets that Choice
AI agents are reshaping marketing, delivering 3‑5% annual productivity gains and accounting for more than 60% of AI‑generated value, according to McKinsey. The technology could add up to $4.4 trillion in global productivity, with marketing alone capturing roughly $463 billion. However, the...
How to Make AI Work for Britain: Consolidate Demand, Diversify Supply
The UK public sector is accelerating AI deployments but faces a “silent lock‑in” risk as fragmented procurement creates mismatched infrastructure and governance. Alan Brown’s new report, “Making AI work for Britain,” proposes a two‑pronged strategy—consolidate demand and diversify supply—to develop...
OpenAI Misses Revenue, User Targets As CFO Fears $1.5 Trillion In Commitments Can't Be Paid
OpenAI disclosed that it missed both its user‑growth and revenue targets for the latest quarter, prompting CFO Sarah Friar to warn that the company may not be able to honor its $1.5 trillion compute‑spending commitments. The shortfall has triggered board scrutiny...
Mindful Optimism: AI's Economic Impact Explored
I did this interview with Alexandra Tremayne-Pengelly where we discussed some of the backstory on my lifelong quest to understand how AI is transforming the economy and why I'm a "mindful optimist". See first comment for the link.
AI‑crafted Fake Nurse Dupes Thousands, Exposing Trust Gaps
The "Emily Hart" story is wild. A medical student in India used AI to create this persona, a patriotic nurse, to grift thousands of people. Why do we fall for it?
Ambient Computing Is (Almost) Here, but Are We Ready for a Post-Product Era?
Ambient computing—technology that anticipates user needs and embeds intelligence in environments—is moving from niche IoT devices to a pervasive, post‑product ecosystem. The article outlines how sectors from FMCG to healthcare will shift from selling standalone products to orchestrating context‑aware services,...
When Human in the Loop Becomes a Legal Fiction
Enterprises tout "human in the loop" as a safety net, but the phrase often masks a rubber‑stamp process that offers little real control. The article argues that true oversight requires three elements—context, authority, and tolerance for delay—and warns that without...
Seoul Viosys Gearing Up for $60B AI Data Center Optical Communications Market
Seoul Viosys is leveraging its proprietary “No‑wire, No‑package” patents and recent VCSEL acquisition to enter the fast‑growing AI‑driven data‑center optical market, estimated at $60 billion. The company has completed a 100 G (25 G×4) transceiver design and is shifting from component supply to...
Microsoft, OpenAI Change Contract Terms — Again
Microsoft and OpenAI announced a revised partnership that drops the exclusivity clause and changes revenue‑sharing terms. OpenAI can now deliver its models on any cloud, while Microsoft retains a non‑exclusive license to the IP through 2032 and will no longer...
ICSC, McKinsey AI Retail Report Forecasts $1 Trillion US Sales by 2030
ICSC and McKinsey & Company released a joint report that projects up to $1 trillion in U.S. B2C retail revenue from AI‑enabled commerce by 2030. Based on a 3,004‑consumer survey, the study splits future store success into convenience‑focused and discovery‑driven formats,...
DG Matrix Appoints Former Microsoft Data‑center Chief Christian Belady to Executive Advisory Board
DG Matrix announced the addition of Christian Belady, the former Microsoft Vice President of Data Center R&D and originator of the Power Usage Effectiveness metric, to its newly formed Executive Advisory Board and as a personal advisor to CEO Haroon...
Philo Elevates Laura Dechant to COO to Accelerate Scaling and AI Integration
Philo announced Laura Dechant's promotion to chief operating officer, adding to a leadership overhaul that also includes Tom Tidgwell as CFO. The move is designed to sharpen operational agility, expand AI integration, and support the company's growth among price‑conscious viewers.
Harvard Study Shows AI Slashes Junior Lawyer Task Time, Puts Partner Role in Question
Harvard Law School researchers report that 58% of attorneys using AI tools complete work faster, rising to 65% among premium users. The speed gains are offset by a perceived gap in deep legal reasoning, prompting senior partners to reassess their...
Illumend CEO Calls for Compliance‑First AI in COI Tracking
illumend chief executive Kristen Nunery told industry leaders on April 27 that AI‑powered COI tracking should be measured by its ability to improve compliance decisions, not merely by how fast it processes documents. Her comments highlight a growing split between...
Workday Shifts to AI‑Agent Platform, Leveraging $3 B in Acquisitions
Workday announced a strategic transformation from a traditional system‑of‑record to an AI‑agent platform, underpinned by almost $3 billion in acquisitions of HiredScore, Evisort, Paradox and Sana. The move follows co‑founder Aneel Bhusri’s return as CEO and a new executive lineup aimed...
Magnificent 7 Set to Unveil Q1 2026 Results, AI Monetization and Leadership Shifts in Focus
The five Magnificent 7 mega‑cap tech firms—Amazon, Alphabet, Microsoft, Meta and Apple—are slated to release Q1 2026 earnings this week. Analysts project a 22% revenue jump for the group, with Alphabet expected to post $106.97 billion in sales and $2.73 EPS, while...
Gen Re Warns Generative AI Could Drive US Insurance Fraud to $40 Billion by 2027
Reinsurance giant Gen Re, in a study with the National Insurance Crime Bureau, warned that generative AI is making false claims cheaper and harder to detect, potentially inflating U.S. insurance fraud losses from $12.3 billion in 2023 to $40 billion by 2027....
CIO 100 Leadership Live Los Angeles Tackles the AI Execution Gap
The CIO 100 Leadership Live conference in Los Angeles brought together senior IT leaders to address the AI execution gap, emphasizing misalignment, fragmented ownership, and cultural challenges. Speakers urged CIOs to reshape structures and leadership moments to turn AI pilots...
Microsoft Launches AI Max, Embedding Ads Directly in Copilot Answers
Microsoft Advertising unveiled AI Max for Search at its Spring Summit, a tool that inserts product callouts into Copilot answers. Early pilots report a 5% lift in click‑through rates, signaling a shift from keyword‑based to conversational ad targeting.
PharosIQ Launches atlasIQ Intelligence at Forrester B2B Summit, Targeting Real-Time Buyer Insight
pharosIQ debuted its atlasIQ Intelligence platform at the Forrester B2B Summit in Phoenix, positioning the solution as the next‑generation buyer‑intelligence layer for go‑to‑market teams. The launch coincides with the company's reported double‑digit organic revenue growth in 2025 and continued momentum...
The Wire by Acutus Found to Be Mostly AI‑Generated, Raising Credibility Concerns
A Mashable investigation uncovered that The Wire by Acutus, a news outlet launched in late 2025, publishes almost entirely AI‑generated content. The finding fuels worries about transparency, misinformation and the need for regulatory oversight of AI‑driven media.
RIL to Set up India’s Largest Data Centre Cluster in Andhra Pradesh
Reliance Industries announced a ₹1.6 lakh crore (~$17 billion) investment to build India’s largest data‑centre cluster in Visakhapatnam, Andhra Pradesh. The giga‑scale AI facility will total 1.5 GW of capacity across three phases, with a 500 MW first phase slated for commercial operation by October 2028 and...
Genie Becomes Databricks’ Core AI for Semantic Analysis
𝐆𝐞𝐧𝐢𝐞 is now the most important way to do data analysis in Databricks. What's unique about it is its ability to extract semantics from your entire Lakehouse, enabling it to answer complex data questions that cripple agents without a deep...
True, OYMotion Pioneer Neuro AI Technology
True Corporation’s Research and Innovation Centre has teamed with neuro‑tech firm OYMotion to create a neuro‑AI platform that translates brain signals into commands for smart devices and prosthetics, aiming to accelerate rehabilitation. The system combines brain‑computer interface sensors, AI‑driven signal...
Vintage LLM Delivers 1930s Siri‑Style Sushi Orders
The new LLM trained only on pre-1931 text is small enough that it can potentially run on device, so, with the right tools, you can get a fully vintage version of Siri, but from the era of Downton Abbey. Here, I...
Will China Restrict US Investment Into Its Top AI Startups?, Anthropic Creates a Hub In Singapore
China is reportedly preparing to block U.S. investment in its leading AI startups unless regulators approve, mirroring U.S. Treasury rules that take effect in January 2025 to curb American capital flowing into Chinese AI, semiconductor and quantum firms. The move...
77% of IT Managers Say Their AI Agents Are Out of Control - 5 Ways to Rein in Yours
A Rubrik ZeroLabs survey reveals that only 23% of IT managers feel they have complete control over enterprise AI agents, while 81% report that agents demand more manual auditing than they save. The study shows 86% of respondents expect agent...
Australian Gov’t Signs AI Agreement with Microsoft Under National Plan
The Australian government has signed a memorandum of understanding with Microsoft, marking the second AI‑focused agreement under its National AI Plan. The non‑binding MoU sets out cooperation on AI safety, workforce upskilling, and cloud infrastructure, while highlighting Microsoft’s $5 billion commitment...
Re‑Architecting Capability for AI: Governance, SMEs, and the Talent Pipeline Paradox
Enterprises are accelerating AI adoption, yet most lack the governance, architectural capability, and leadership readiness to scale AI responsibly. The paper argues that human‑led governance, subject‑matter experts (SMEs), and experience‑based judgment are non‑negotiable pillars for sustainable AI integration within enterprise...
212. America's Semiconductor Policy and the AI Race with China
In this episode, host Chad Bowne and guest Dan Kim—former chief economist of the U.S. Commerce Department’s CHIPS Program and veteran of Qualcomm and SK Hynix—explore America’s aggressive semiconductor policy amid the AI race with China. They trace the historical...
AI Is Powering the Next Wave of Consumer Robotics
I found myself in a panel with Sebastian Thrun who started the Waymo company. I couldn’t help but record him for a few seconds. I was wearing the @Looki_ai pendant. Will be interesting to see what it does tonight. Talking about how...
Claude Beats ChatGPT and Gemini in Design Quality
Just gave Claude (Opus 4.7 and Sonnet 4.6), ChatGPT 5.5, and Gemini 3.1 the same prompt to build a deck design. Gemini & ChatGPT outputs were intern-level at best. Claude was senior designer. Crushed it. Not surprised. Kept prompt simple:...
Big Tech AI Spending Binge Results in Massive Job Cuts!
Big tech firms are redirecting massive capital toward AI infrastructure, with projected 2026 capex of $674 billion—more than double 2024 levels. To fund the spend, hyperscalers, SaaS vendors and telecom providers are slashing headcounts, exemplified by Meta’s 8,000 layoffs and Microsoft’s...
AI Robot Enhances Guest Experience with Real-Time Guidance
#AI Meets Hospitality: Realbotix #Robot Greets and Guides Visitors by @CyberRobooo #Robotics #ArtificialIntelligence #Innovation #Technology https://t.co/h3IhbYiWo7
Foundry Powers Durable, Stateful Agents for Continuous Workflows
Great example of what Foundry enables: durable, stateful agents that run across time boundaries, orchestrate tools and models, and close the loop with evaluation and improvement over long-running workflows. @jeffhollan https://t.co/v0yWTDIcn3
Claude Needs 15 Minutes to Convert HTML Into PDF
This cant be true. It takes Claude 15 minutes to convert an HTML to a pdf.
AI's Collective Roots Demand Shared Prosperity and UBI
#AI was built on all our labor. Every word, image, idea came from us. Its gains belong to everyone, not just a few. I signed the AI Pledge for Humanity, committing 2% of my income to Income to Support All Foundation...
Agentic AI Boosts BMW's Engineering Speed Twelvefold
The AI-powered R&D department: how agentic AI is supercharging engineering velocity “BMW accelerates analysis cycles twelvefold” https://t.co/wshZeLW7q3 https://t.co/6Sl6SlBiic
Google Gemini's Lyria Generates Full Songs, Not Just Lyrics
Creating song lyrics with AI is so 2025. Now you can compose entire songs using the Lyria model in Google Gemini. It's pretty darn astonishing. Here's how it works, with sample tracks in a variety of styles: https://t.co/7pNLIXLd7b #gemini #lyria...
AI Adoption Outpaces Enterprise Readiness, Creating Chaos
It’s absolutely insane how fast AI is getting adopted when it’s nowhere near the adoption stage for a large companies. What a nightmare.