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
The Honest AI Onboarding Curve Nobody Tells You About
Small businesses adopting AI agents face a steep onboarding curve that initially hurts productivity. In weeks one and two, output quality drops and speed slows as the agent learns company‑specific workflows, often achieving only about 60% accuracy. By week four, the agent can match or exceed manual quality and operate up to ten times faster. However, only roughly 15% of AI pilots transition to full production because most firms lack the organizational infrastructure to scale the technology.
Opus 4.7 Uncovers Hidden Health and Music Patterns
Opus 4.7 with Apple Health is wildly fun. So far it’s guessed my social activity from noise levels. It noticed an asymmetry in my walk during a certain week (had knee pain). It’s asking why I stopped listening to music a...
US Nuke Scientists Are Mysteriously Disappearing! - New World Next Week
In this episode of New World Next Week, hosts James Corbett and James Palato discuss Anthropic’s AI model Mythos, a zero‑day vulnerability generator that’s being quietly tested by major tech firms under “Project Glass Wing,” and the unsettling pattern of...
Deconstructing the Mythos Myth
In 2019 OpenAI chose not to release its 1.5‑billion‑parameter GPT‑2 model, citing fears it could generate deceptive, biased or abusive content at scale. Instead the lab offered a less powerful version to researchers while publishing extensive technical details. The decision...
AI Does McKinsey‑grade Research, Drives Millions per Lead
You can run McKinsey-level research for almost nothing to generate millions in revenue. Here's what our Single Brain actually generates for every single lead that books a discovery call - before anyone touches it: 1. Full company dossier: revenue, headcount, tech stack,...
Mozilla Throws Thunderbolt at Enterprise AI Providers
Mozilla’s MZLA subsidiary unveiled Thunderbolt, an open‑source AI client aimed at enterprises seeking data‑sovereignty. The client integrates with deepset’s Haystack platform and supports Model Context Protocol and Agent Client Protocol standards, allowing firms to run any LLM on‑premise or in...
Unintentional AI Adoption Is Already Inside Your Company. The Only Question Is Whether You Know It.
In‑house counsel are confronting AI that has already seeped into daily workflows through browsers, email extensions, and personal devices, often without any formal approval. This unintentional adoption creates a hidden layer of corporate memory built from prompts, scraped emails, and...
Discover Matt's Cutting-Edge Research at Stanford Seminar
It will be great learn about Matt 's latest work at our Stanford Digital Economy Lab lunch seminar. I first got to know Matt back at the Massachusetts Institute of Technology when he was doing his PhD and he's continued...
Gain Full Visibility Into AI Spend with Ramp
Anthropic made its first dollar three years ago. last month it crossed $30B in revenue. that money is coming from somewhere, and your CFO probably can't tell you where. the problem isn't the spending. the companies on Ramp investing the most...
OpenAI Starts Offering a Biology-Tuned LLM
OpenAI unveiled GPT‑Rosalind, a large language model fine‑tuned for biology workflows. Trained on 50 common biological tasks and public databases, it can suggest pathways, prioritize drug targets, and connect genotype to phenotype. The model is deliberately more skeptical to curb...
Army’s Work With OPV Black Hawk To Inform Bringing Autonomy To MV-75 FLRAA
The U.S. Army has accepted the first H‑60Mx Optimally Piloted Vehicle (OPV) Black Hawk, equipped with Sikorsky’s MATRIX autonomy suite, to serve as a testbed for autonomous capabilities on the upcoming MV‑75 Cheyenne II FLRAA tilt‑rotor. Recent DARPA‑originated demonstrations, including the...
Fewer Federal Workers, Same Mission: Why AI Is the Productivity-First Technology Critical to Agency Operations and Efficiency
The federal government is confronting a shrinking workforce—317,000 employees departed in 2025—while still tasked with delivering services. Budget pressures have limited overall spending cuts, but IT outlays rose to $126 billion in 2024, prompting agencies to turn to artificial intelligence for...
New Guidelines Recommend AI-Based Breast Cancer Risk Assessments
The National Comprehensive Cancer Network’s 2026 Clinical Practice Guidelines now recommend image‑based artificial‑intelligence risk assessments as a primary tool for breast cancer screening. The guidance advises using AI‑derived five‑year risk scores from routine mammograms, with a 1.7 % risk threshold prompting...
Lack of Governance Coordination on AI Costing Companies
Grant Thornton’s 2026 AI Impact Survey reveals that weak governance and compliance barriers are the top reason—cited by 46% of respondents—AI projects fail or underperform. Although 75% of organizations continue to approve major AI spend, nearly half have not set...
Blue Owl Builds a Capital Platform for the Hyperscale AI Era
Blue Owl Capital has transformed from a real‑estate investor into a dedicated capital platform for hyperscale AI data‑center projects. The firm integrated IPI Partners in January, adding over $11 billion in assets and appointing a digital‑infrastructure head. In May, its Digital...
Context Switching Costs More Than Software—AI OS Solves It
The most expensive “tool” in a solo business isn’t software. It’s context switching. Every switch from “write a post” → “check Stripe” → “update launch board” costs real money in dropped threads and half-finished work. That’s the quiet tax that AI OS tools...
From Search to Conversation: How Is AI Rewiring the Traditional Shopper Funnel?
In this Euromonitor webinar, experts explain how AI is fundamentally reshaping the traditional shopper funnel, moving from keyword‑driven search to algorithmic, conversational discovery. They highlight early disruption signals such as TikTok Shop’s influencer‑driven model and the explosive growth of ChatGPT,...
Val Kilmer’s AI Movie Trailer Reveals Major Problem With AI Actors
The upcoming drama "As Deep as the Grave" uses generative AI to insert a digital Val Kilmer, who died in 2025, into a role he never filmed. The trailer, released with the consent of Kilmer’s estate, shows an uncanny, poorly rendered...
Opus 4.7 Generates Surprisingly Impressive Unicorn Sketch
On the plus side with Opus 4.7, if it does decide to think it produces BY FAR the best Sparks unicorn* ever, even non-thinking is pretty good, if not great. * This is created using TikZ, which is a language built...
Waymo Matches Human Drivers' Safety, Despite Underreported Crashes
Waymo in SF has comparable safety rate to UberLyft human-driven rides in SF, says Cardozo prof Matthew Wansley, citing academic study, but notes the human-driven crashes in this set are likely underreported. He thinks AVs are making a “positive contribution”...
The Tactical Edge Is Now: Deploying AI and Communications in Disconnected Environments
Federal IT modernization has long prioritized cloud‑centric, multi‑cloud solutions, but field operations often lose connectivity, forcing a shift to resilient edge architectures. The article argues that graceful degradation—designing systems to shed complexity and keep core functions alive—is the defining metric...
AI Labs Purchase Defunct Startup Communications, Raising Privacy Concerns
AI labs are buying internal communications of defunct startups to train their agents. Emails, Slack archives, etc. Personally identifiable info is removed by data resellers. But how would you feel knowing your former board/CEO is selling your comms to recover...
Opus 4.7 Improves, yet Anthropic Releases While Mythos Stays Hidden
Opus 4.7 is a big improvement...but also Anthropic is ok releasing it but not Mythos? I'm so confused... https://t.co/OVBq4ZWKyH
Overheard At HumanX 2026
At HumanX 2026, leading technologists and policymakers warned that AI will soon integrate directly with human cognition, making it hard to distinguish machine‑generated thoughts. They highlighted the rise of digital twins for enterprise simulation, especially in mergers and acquisitions, and...
Earbuds with Cameras Let AI See Your World
Tiny cameras in earbuds let users talk with #AI about what they see by Stefan Milne @TechXplore_com Learn more: https://t.co/CPoubUQApt #MachineLearning #ArtificialIntelligence #ML https://t.co/Ih6hAH3R3A
Agents Degrade Like Computers—Restart or Refactor?
I'm starting to realize Agents are kinda like when you get a new computer: it starts fast, clean, organized...and then over time it becomes slow, bloated, and difficult to use. Does that mean I just start over? Or try to...
How Controlling Light Inside a Tiny Resonator Could Speed AI Chips and Secure Communications
KAIST researchers unveiled a dual‑bus integrated photonic resonator that can precisely shape the spectrum and phase of light, overcoming the limitations of traditional single‑bus designs. The device enables engineered interference, allowing optical signals to be customized for high‑performance computing. Led...
New Method Trains Leaner, Faster AI Models On‑the‑Fly
New technique makes AI models leaner and faster while they’re still learning https://t.co/6XKTYsxe3N #ArtificialIntelligence #Innovation #Technology #Tech #TechNews https://t.co/eRLguCAQMw
Kennedy: 90% Of FDA Reviewers Are Using AI For Faster Drug Approvals
U.S. Health and Human Services Secretary Robert F. Kennedy Jr. told the House Ways & Means Committee that more than 90% of FDA reviewers are now using artificial‑intelligence tools to speed drug approvals. The AI applications are also being rolled...
FDA Clears First AI-Enabled, Detector-Based Spectral CT System
Philips received FDA 510(k) clearance for its Verida spectral CT system, the first AI‑enabled, detector‑based scanner of its kind. The platform combines a dual‑layer Nano‑panel detector with AI‑driven image reconstruction, delivering always‑on spectral imaging without extra scans. Verida can reconstruct...
Tesla's Self-Driving 14.3.1 Navigates SF Like AGI
Tesla Self-Driving 14.3.1 tours San Francisco. Sometimes it feels like driving through this city requires AGI, but Cybertruck just does it so perfectly. https://t.co/Rayj1Bzi7G
Real-Time Revolution: How AI Is Changing the Deposition Game with Dean Whalen & John Skelton
In the Red Cave Law podcast, Dean Whalen of Readback (InforWare) and John Skelton of Seyfarth Shaw discuss AI‑driven deposition software that delivers real‑time, searchable transcripts. The platform uses a proprietary speech‑to‑text engine hosted in a secure environment, offering instant...
Anthropic Debuts ‘Repeatable Routines’ in Major Claude Code Automation Update
Anthropic rolled out a major Claude Code update that adds “repeatable routines,” an AI‑driven automation feature, and a redesigned desktop IDE. Routines let developers schedule or trigger tasks—such as API calls, GitHub actions, or maintenance jobs—directly within Claude, running on...
Anthropic Squeezes Enterprises by Ejecting Bundled Tokens From Seat Deal
Anthropic has overhauled its enterprise seat pricing, replacing the former $20‑per‑employee plan that bundled token allowances with a flat monthly fee that no longer includes any usage credits. All token consumption is now billed at standard API rates, effectively turning...
AI Agent Delegation via MCP Has Gaps a Murderbot Could Walk Through
Anthropic’s Model Context Protocol (MCP) expands data‑sharing among AI agents, but securing those interactions remains a challenge. At the 2026 MCP Dev Summit, Gluu CEO Michael Schwartz warned that relying on a single gateway for zero‑trust is insufficient and advocated...
Nvidia's $2 B Marvell Deal Highlights AI Infrastructure Surge and a Quiet Edge Play
Nvidia poured $2 billion into Marvell Technology to embed its NVLink Fusion interconnects in data‑center fabrics, cementing a high‑bandwidth AI backbone. The partnership also shines a light on Nokia, which is set to receive a $1 billion Nvidia investment for its AI‑RAN...
How to Create AI Agents for Social Media Marketing
Social media teams are overwhelmed by the volume of platforms and messages, prompting a shift toward AI agents that can autonomously generate content, monitor trends, and handle customer interactions. The guide outlines how to build such agents—from selecting LLMs and...
Free ChatGPT Downgrade Reveals Unexpected Limitations
I downgraded my ChatGPT to free a few weeks ago as an experiment, and look what I just got https://t.co/syuUNh3CE4
CenterSeat Elevates Founding Engineer Saurav Mishra to CTO to Accelerate AI‑Security Platform
CenterSeat announced the promotion of founding engineer Saurav Mishra to chief technology officer. Mishra will now steer technology strategy, product architecture, and engineering as the Austin‑based startup expands its AI‑driven compliance platform. The move underscores CenterSeat’s focus on scaling its...
PwC to Merge Risk and Consulting Units in Global Overhaul Driven by AI
PwC is overhauling its global consulting model by merging its risk and consulting divisions, a move aimed at standardising services and leveraging AI. The blueprint, drafted by senior leaders in the US and UK, seeks to address fragmented delivery and...
Docupace Names Mike Conlon Chief Revenue Officer to Accelerate Wealth‑Tech Growth
Docupace announced the appointment of Mike Conlon as Chief Revenue Officer. The veteran SaaS leader will steer a unified go‑to‑market strategy as the firm expands its AI‑driven back‑office suite for wealth‑management firms. The hire signals a push to scale revenue...
OnePlan Launches AI‑enhanced April 2026 Release to Automate Enterprise Portfolio Planning
OnePlan announced its April 2026 product release, introducing AI‑powered plan automations, bulk CSV import, and new financial dashboards. The upgrades aim to cut manual effort and boost visibility for the 500+ companies that use the platform.
Freshfields Deploys Google Gemini AI to 5,000 Lawyers, Scaling Generative Tools Firmwide
Freshfields announced that more than 5,000 of its lawyers and staff are now using generative AI tools built on Google Gemini, a year after launching a strategic partnership with Google Cloud. The rollout includes bespoke legal platforms, NotebookLM Enterprise for...
Phenom CEO Says AI Is Killing Traditional SaaS at 2026 Conference
Phenom co‑founder and CEO Mahe Bayireddi told attendees at the company's 2026 conference that AI is making software cheap and that the traditional SaaS model is dying. The warning underscores a shift toward data‑centric value in HR technology and could...
OpenAI Pulls Out of Second Stargate Data‑center Deal, Handing Norway Site to Microsoft
OpenAI has withdrawn from a second Stargate data‑center partnership, abandoning the UK project and allowing Microsoft to take over the Norway Narvik facility. The move underscores a strategic shift in OpenAI’s compute sourcing amid mounting costs and partner tensions.
Allbirds Shares Soar 600% as Shoe Maker Rebrands to NewBird AI, Launches $50M GPU‑as‑Service Push
Allbirds Inc. announced a $50 million convertible financing facility and the sale of its footwear business for $39 million, then rebranded as NewBird AI to sell GPU compute capacity to enterprises. The surprise pivot sent the Nasdaq‑listed stock up nearly 600% in...
Atlassian Launches Agentic Pipelines in Bitbucket to Automate Dev‑ops Chores
Atlassian announced the rollout of Agentic Pipelines, an AI‑enabled extension to Bitbucket Pipelines that automates repetitive pre‑ and post‑code tasks such as documentation updates. The feature aims to reclaim the 84% of developers' day spent on non‑feature work, promising faster...
Cellulant Names Anthony Hernandez COO to Accelerate AI‑Powered Payments Ops
Cellulant announced the appointment of Anthony Hernandez as chief operating officer, tasking him with building an AI‑enabled, data‑driven operating model for its pan‑African payment platform. The move signals a strategic push toward real‑time compliance, operational transparency and scalable growth for...
Amazon Launches Ember Artline TV, a $900 Rival to Samsung’s Frame
Amazon has opened pre‑orders for the Ember Artline, a 4K QLED TV priced at $899 for the 55‑inch model and $1,099 for the 65‑inch version. The TV, billed as a lifestyle product, offers a matte screen, free access to 2,000...
Global InsurTech Funding Plummets 78% in March, Hitting 2026 Low
Global InsurTech funding fell 78% in March 2026, slipping below $420 million and marking the lowest level of the year. The only mega‑round was Paris‑based Alan’s €100 million raise, while most deals focused on AI and embedded insurance infrastructure.