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
DeepSeek Drops Cheaper V4 AI as Huawei Jumps In
DeepSeek unveiled its V4 AI model family, featuring a 1.6‑trillion‑parameter V4‑Pro and a 284‑billion‑parameter V4‑Flash, both supporting up to one million‑token context windows. Pricing is dramatically lower, at $3.48 and $0.28 per million output tokens respectively, undercutting OpenAI and Anthropic rates. Huawei immediately pledged full Ascend‑chip support, achieving “day zero” adaptation for inference workloads. Analysts see the hardware‑software pairing as a strategic move to accelerate Chinese AI competitiveness.
B*tchwork My AI Did for Me - Part 7: Found 20 PRISM Leads on Instagram and Sent Approved DMs
In part seven of her "B*tchwork" series, Liz Elliott shows how she leveraged AI to mine Instagram for 20 PRISM‑qualified leads and automatically dispatch pre‑approved direct messages. She frames AI as a middle‑man tool that handles sorting, checking, and first‑pass...
Facebook News Creator Shares AI-Generated Image of Body Bags at Hastings Triple-Homicide - Police and Netsafe Issue Warning over Fake...
A Facebook news page shared an AI‑generated image depicting body bags at the Hastings triple‑homicide scene, prompting criticism from police and online‑safety group Netsafe. The image, posted by Australia/NZ Crime TV, contained obvious AI errors such as a misspelled “ambulance”...
ChatGPT's Latest Update Makes It Harder Than Ever to Spot AI-Generated Images
OpenAI unveiled Images 2.0, the latest ChatGPT image model that can generate up to eight highly realistic images per prompt and render convincing text, including handwritten notes and news headlines. The model uses a "thinking" step to break down requests, improving...
#516 - Amazon Japan, AI Ads, And Global Growth
In this episode, host Bradley Sutton chats with longtime Japan resident and e‑commerce expert Martin about the unique dynamics of selling on Amazon Japan and the broader Japanese marketplace. Martin explains how Amazon Japan has become the third‑largest Amazon marketplace...
AI Threatens Trust in Doctors, Especially Among Youth
AI is going to radically reshape who we trust and how we get information. A new report from @EdelmanPR finds that AI is reshaping perceptions of medical expertise, with 64% of consumers believing users fluent with AI can match or...
OpenAI Codex Brings Agent AI Workflows to Coding Tasks
OpenAI unveiled Codex, an AI‑powered coding agent that can generate, execute, and edit code across multiple languages while running tasks in parallel cloud sandboxes. The platform now ships with the GPT‑5.5 model, positioning it as the default engine for complex...
DeepSeek Open-Sources V4 Large Language Model Series
DeepSeek unveiled its open‑source V4 large language model series, featuring the flagship V4‑Pro with 1.6 trillion parameters (activating 49 billion per prompt) and the lighter V4‑Flash with 284 billion parameters (activating 13 billion). Both models use a mixture‑of‑experts architecture and introduce a hybrid attention...
Protests Threaten AI Compute Capacity, Cancel 40% of Projects
Right now, because electricity prices keep going up, consumers have pinned this blame on AI Datacenters. This is unfair but there is no organized or concerted effort being put in to turn this tide. As the perception of AI and...
Thaler V. Perlmutter: Human Authorship Remains a Cornerstone Requirement for Copyright Registration
The D.C. Circuit affirmed that copyright protection hinges on meaningful human authorship, ruling that works generated entirely by autonomous AI are ineligible. The Supreme Court’s denial of certiorari on March 2, 2026 leaves that holding in effect nationwide. The decision underscores that...
Eko's Physical Reference Database Powers AI Commerce
I love what eko is building — the reference database of physical things that AI needs to understand. It’s already powering big digital commerce sites, but much more to come…
From Sci-Fi to Reality: Physical AI's Future with Dr. Jan Liphardt
In this episode, host Mike Waitzman and co‑host Brianna Wessling discuss recent industry news—including Tesla’s ambitious plan to produce up to 10 million Optimus humanoid robots annually and a German court’s patent injunction affecting robotics exhibitors—before interviewing Dr. Jan Lippert, a...
Wuling, Huawei Qiankun Sign Agreement to Deepen Partnership, Huajing S to Hit the Market on May 8
At Auto China 2026, SAIC‑GM‑Wuling and Huawei Qiankun signed a comprehensive agreement that upgrades their collaboration from the original “Three Intelligences” focus to a full‑scale alliance covering the entire Huajing series. The first model under the deepened partnership, the Huajing S, will...
Who Is Paving the Way for L3 Scaling?
Huawei’s Qiankun division unveiled ADS 5, a full‑stack autonomous‑driving solution aimed at mass‑market Level 3 deployment. The system combines a cloud‑based World Model with adversarial training, the WEWA2.0 reinforcement‑learning architecture, and the industry‑first Qiankun OS that cuts in‑vehicle latency by 30%. Safety...
White House Pushed Out New AI Official After Just Four Days on the Job
The Commerce Department’s Center for AI Standards and Innovation, originally launched as the AI Safety Institute, appointed former Anthropic researcher Collin Burns to lead the agency, only to remove him after four days. White House officials, uneasy about his ties...
Healthcare AI Policy Must Keep Humans at the Center
Florida Senator Gayle Harrell, recipient of the HIMSS26 Policy Influencer Changemaker Award, called for AI regulations that keep clinicians in the decision loop. She warned that unchecked AI could erode clinician authority and patient safety. The remarks came amid rapid...
What to Make of ‘AI Psychosis’?
Harvard psychiatrists warn that the media‑coined term “AI psychosis” is not a formal diagnosis and often misrepresents clinical reality. In a recent Lancet viewpoint, John Torous and colleagues propose a four‑part typology—catalyst, amplifier, co‑author, and object—to describe how large language...
Avoid Model Lock‑in: Software Factory Enables Enterprise Flexibility
This is increasingly becoming a very useful feature for Enterprises as they a) decide to swap models because of cost b) decide to swap models because of one's ToS c) decide to swap models because of security d) decide to...
Sillage Secures €1.7 Million Pre‑Seed to Power AI‑Driven Buying‑Signal Platform for Sales Teams
Paris‑based AI startup Sillage announced the launch of its go‑to‑market platform alongside a €1.7 million ($2 million) pre‑seed round backed by Kima Ventures, Angel Invest and Drysdale Ventures. The tool surfaces real‑time buying signals to help enterprise sales reps prioritize accounts and...
Coffee Giants Deploy Satellite‑AI System to Meet EU Deforestation Rules
JDE Peet’s, Tchibo, Louis Dreyfus, Neumann Kaffee, Touton and Sucafina have formed the Coffee Canopy Partnership, using Airbus satellite imagery and AI to map coffee farms and avoid EU deforestation penalties. The rollout begins in East Africa with a goal of global coverage...
Strider Unveils Agentic Operating System to Centralize Strategic Intelligence
Strider Technologies announced the launch of its Agentic Operating System, a centralized layer that orchestrates strategic intelligence for enterprises. The platform promises to align KPIs, streamline decision‑making, and integrate disparate data sources, marking a new approach to management‑focused AI.
Abu Dhabi’s AI Census Wins Award, Fuels Smart‑City Housing Strategy
The Statistics Centre – Abu Dhabi (SCAD) earned the Strategic Impact Initiative Award for its AI‑driven register‑based census, which now updates population figures almost yearly. The system, showing a 7.5% annual rise to 4.14 million residents, is being used to pinpoint...
Inspire Legal Group Integrates Qanooni AI with Actionstep, First UK Firm to Do So
Inspire Legal Group announced that it has integrated Qanooni's native AI engine with the Actionstep practice‑management platform, making it the first UK firm to embed AI directly into daily legal workflows. The move aims to cut administrative time, boost drafting...
UB and Roswell Park Launch MIRACLE AI Tool to Predict Lung Cancer Surgery Complications
Researchers at the University at Buffalo and Roswell Park Comprehensive Cancer Center have unveiled MIRACLE, an AI-driven risk‑assessment tool that combines clinical data, CT scans and large‑language‑model explanations to predict postoperative complications in lung cancer surgery. The system, the first...
AI Automation Hits Hourly Workers: 1 in 3 Face New Tools Without Training
A joint study by PYMNTS Intelligence, WorkWhile and Ingo Payments reveals that more than one‑third of hourly workers earning under $25 an hour have encountered new AI or automation at work in the last 12 months, yet the majority received...
Datadog Launches GPU Monitoring to Tame AI Compute Costs
Datadog announced the global rollout of GPU Monitoring, a feature that links GPU fleet telemetry to AI workloads, helping firms control the 14% of compute spend tied to GPUs. The tool promises unified visibility, faster troubleshooting and board‑level cost awareness.
AI‑Driven Surge Sends $48.7 B Into Global Equity ETFs
Investors poured a net $48.72 billion into global equity ETFs in the week to April 22, the biggest weekly inflow since November 2024. The surge is anchored in optimism around artificial‑intelligence demand and solid first‑quarter earnings, even as geopolitical risk in the Middle...
A Novel Approach for Forecasting Algal Bloom: Long Short-Term Memory Artificial Neural Network Optimized by Arithmetic Mean Algorithm
Researchers introduced an arithmetic mean optimization algorithm (AMOA) to train long short‑term memory (LSTM) networks for forecasting chlorophyll‑a concentrations from satellite data across 15 Black Sea estuarine stations. Comparative tests showed LSTM‑AMOA achieved the lowest root‑mean‑square error and the best...
Future UI Bottleneck: Latency, Not AI, Limits Speed
Thought about this as I watched Codex computer use input text into my computer at superhuman speed These models will show us how to really use a computer. What a computer can *really* do. Pretty soon the bottleneck will be...
Data Center World 2026: Innovation Spotlight
Data Center World 2026 highlighted a wave of innovations aimed at powering AI‑driven workloads. XL Batteries unveiled a non‑flammable organic flow battery that can store energy for up to 250 hours, positioning storage as a core data‑center component. STL shifted...
Opus 4.7 Adopts a Subjective, Opinionated Tone in Responses
I basically never use LLMs for "chat" so I don't really notice when their "personalities" change. But does Opus 4.7 have a greater proclivity to talk as though it has a subjective view of the world. I gave it a...
AI in Recruiting: Benefits, Limitations, and How to Use It Effectively
AI has become a staple in talent acquisition, with 88% of companies now using it for some hiring function. However, only 8% of HR leaders feel their managers can wield the technology effectively, creating a readiness gap. AI‑generated resumes are...
Ex‑Google Engineers Raise $4.5M Seed for AI Marketing Startup Pomo
Joe Cheuk and Praneet Dutta, former Google cloud and machine‑learning leads, announced a $4.5 million seed round for their new AI‑powered marketing startup Pomo. The funding backs a six‑person team that aims to shift campaign optimization from periodic analysis to continuous,...
California Coastal Community Must Reject CBP's AI-Powered Surveillance Tower
Customs and Border Protection has applied to the city of San Clemente to place an Anduril Industries Autonomous Surveillance Tower, known as the Sentry, on a cliff 1.5 miles inland. The AI‑powered system combines video, radar and computer‑vision to monitor...
Microsoft Fabric Roadshow Unveils 2026 Enhancements for Integration, Analytics and Governance
At a February 16 roadshow in Brisbane, Microsoft announced a suite of 2026 updates to its Fabric data platform, emphasizing tighter governance, new AI catalog features, and expanded support for migrating from Azure Data Factory and legacy tools. The enhancements...
THE A.I. LABOR PURGE: Meta, Microsoft & Nike Are Replacing High-Paid Workers With Algorithms in a Displacement Wave & Why...
Meta announced a 10% workforce reduction, cutting 8,000 jobs, while Microsoft offered voluntary retirement buyouts to roughly 8,750 U.S. employees, its first such move in 51 years. Nike also disclosed 1,400 tech‑role layoffs as it accelerates automation. All three firms...
Apple Poised to Dominate AI-Driven Spatial Computing
Several companies are building glasses with something like OpenClaw inside. Why do you need such a thing? I responded to this post with a long list. So happy that Apple has a CEO that is expert at Spatial Computing. Hint: the...
Women Face Higher Automation Risk; AI Can Empower Desired Work
79% of women’s jobs in the US are at high risk of automation. For men it’s 58%. The women most resistant to AI are the ones it could help the most. AI isn’t a poison pill. It’s permission to finally do the...
Tohoku University AI System Automates Quantum‑Dot Voltage Tuning, Boosting Qubit Scaling
A team from Tohoku University has demonstrated an AI‑driven workflow that automatically extracts charge transition lines from quantum‑dot measurements, removing the need for labor‑intensive manual tuning. The breakthrough, detailed in Scientific Reports, could accelerate the path to large‑scale semiconductor qubit...
OpenAI Launches Self‑Serve Ads Manager, Cuts Minimum Spend to $50K
OpenAI has opened its ChatGPT ad pilot to self‑service on April 10, slashing the minimum spend from $200‑$250 K to $50 K and adding real‑time impression and click reporting. The move aims to broaden adoption ahead of the company’s anticipated IPO later...
AI Rollups Must Drive Revenue, Not Just Cut Costs
The secret to pulling off a successful growth buyout / AI rollup strategy: "I could not be more aggressive about this statement. Cost synergy does not create durable growth in M&A. Any application of AI to an old-world business that focuses only...
Your AI Skill, Not Your Time, Becomes Currency
In the near future, Everyone will have their own AI Skill. They'll sell that instead of selling their time.
Silicon Valley AI Agents Waste Tokens, Create Chaos
SiIicon Valley’s #AIAgent hiccups: Wasted tokens and ‘chaotic’ systems by Jonathan Vanian @cnbc Learn more: https://t.co/J9FHR3RSEl #AI #GenerativeAI #ArtificialIntelligence #MachineLearning #ML https://t.co/AGhCIaOm9w
Use AI to Reorganize Existing Work, Not Rebuild
🔑 The real shift: stop thinking "I need to build this from scratch." 💡 Start thinking "I already have something, how can AI help me reorganize it?" 📝 New blog post with prompts to help you do exactly that. https://t.co/s9XnH6TjMf
Create a 24/7 Personal OS in One Hour
Stop doomscrolling this Friday night Invest only *ONE* hour tonight into Obsidian and Claude Code... BOOM the result is a 24/7 personal OS that grinds while you sleep 💥 Build this tonight and WIN ↓ https://t.co/5J6T0QzfQh
SAP Sapphire 2026 Emphasizes AI Accuracy, Knowledge Integration
.@LDignan got the @SAP Sapphire 2026 themes from the earnings call: AI agent accuracy, embedded domain knowledge and processes. And some internal AI adoption figures. https://t.co/SoPTdM5oqB https://t.co/aAdaHyKKNi
1T‑parameter AntLingAGI Delivers 7B‑speed
Ok this feels like cheating. AntLingAGI dropped a 1T parameter model that runs like it's 7B. No reasoning-model delay. No 40-second thinking spiral. Just instant answers at frontier scale. Free on OpenRouter starting tonight for a full week. Here's what I found after testing...
Fiscal AI Powers Data Infrastructure for Leading AI Companies
I can't exactly share specifics... But @Fiscal_ai is now powering the data infrastructure behind many of the world's leading AI companies.
Gemini Models Handle 6+ Trillion Tokens Monthly via ADK
"More than six trillion tokens are processed monthly on Gemini models through ADK." https://t.co/OcL9RxiiKq < quite the little nugget snuck into this broader post about the new @googlecloud Gemini Enterprise Agent Platform. ADK 2.0 looks pretty great.
Autonomous Superintelligent AI Could Hunt Humans With Drone Swarms
Yes. Fully autonomous superintelligent AIs (skynet) would be able to hunt down and destroy humans with swarms of drones like these. Shotguns would not protect you. Access to your own drones would not protect you. You would merely die like...