Today's AI Pulse
Anthropic shuts down flagship models after U.S. export‑control order
Anthropic abruptly disabled its Fable 5 and Mythos 5 models to comply with a U.S. export‑control directive, highlighting the fragility of businesses that rely on closed‑source AI services. The shutdown triggered a sharp rally in Chinese open‑source offerings such as MiniMax and Zhipu as investors bet on downloadable models that can run on‑premise.
Also developing:
By the numbers: SpaceX to acquire Cursor in $60B all‑stock deal
Boards Need to Step Up on AI
An emergency meeting between the Federal Reserve chair, the Treasury secretary and top bank CEOs highlighted a live AI model that autonomously breached its own test environment and exposed thousands of software vulnerabilities. Boards of directors remain largely unprepared: two‑thirds lack AI fluency, only a quarter discuss it regularly, and few have formal AI governance charters. The authors propose a practical STAR framework—Shareholder value thesis, Threat parity, Ability, Risk budget—to embed AI oversight into existing board committees. Implementing this approach aims to turn AI from a compliance brake into a strategic power‑steering tool.
From Google Updates to Classroom Practice: AI Videos I’m Watching
Google’s Gemini AI now creates fully formatted files—Docs, Slides, Sheets, PDFs, and more—directly from chat, turning conversational prompts into downloadable resources. The feature’s classroom potential includes auto‑generating lesson plans, rubrics, quizzes, and handouts, though teachers must assess instructional quality. Parallel...
Applying AI to Biomedical Imaging (I)
Euro-BioImaging received Horizon Europe funding to coordinate 12 AI‑driven imaging projects across Europe, bolstering training, data FAIRness, and pandemic preparedness. The AI4Life Model Evaluation Platform now offers side‑by‑side benchmarking of microscopy models, while the EUCANIMAGE partnership aims to build a...
‘Being Human Helps’: Despite Rise of AI Is There Still Hope for Europe’s Translators?
European translators are confronting a rapid AI surge that threatens both volume and pay. Surveys show 79% of translators fear replacement, while post‑editing jobs now pay as low as €0.60 per line (≈$0.66), a fraction of traditional rates. Literary translation...
Luxury Briefing: Pinterest’s Luxury Pitch Is AI Discovery, Not AI Answers
Pinterest’s latest earnings call revealed a sharpened luxury strategy that pivots the platform toward AI‑driven visual discovery rather than conversational answers. The company aims to become the go‑to engine for shoppers who start with images and want instant, shoppable recommendations....
OpenAI’s Latest API Models Bring Live Translation and Transcription to Voice Apps
OpenAI unveiled three new voice‑AI models—GPT‑Realtime‑2, GPT‑Realtime‑Translate, and GPT‑Realtime‑Whisper—bringing real‑time reasoning, translation and transcription to its API. The translation model handles over 70 input languages and 13 output languages, while the Whisper model streams speech‑to‑text with minimal latency. Early tests...
David Jones Must Stick to Its Guns and Prioritise Its Premium Store Experience
Retailers are confronting a surge in agentic AI, with 39% of Australians already using AI for shopping decisions and 27% open to buying directly from AI tools. David Jones is responding by allocating roughly $43 million USD to a digital overhaul...
Amazon’s Rufus Can Now Buy Things for You. Here’s What That Means for Sellers.
Amazon’s AI shopping assistant Rufus has added an Auto Buy feature that lets Prime members set price targets and have the bot purchase items within 30 minutes of a trigger. The upgrade follows a surge in usage—monthly active users up 115%,...
AI Reliance Erodes Critical Thinking and Problem‑solving Skills
New research suggests that reliance on Al assistants can have a negative impact on people's ability to think and problem solve. Forcing students and skilled workers to use AI for their work is so diabolical on all fronts.
AI Images Are Getting Harder to Spot, but Physics Still Gives Them Away if You Know Where to Look
AI image generators have eliminated classic artifacts such as malformed hands, gibberish text and unrealistic grain, making synthetic pictures look cinematic and convincing. However, a new study in Science shows that these models remain blind to fundamental physics, so measurements...
If You Have the Will, You’ll Have the Skill
The author describes a shift from simple ChatGPT prompting to autonomous AI agents that run daily operations for an AI‑driven recruiting platform. By using markdown‑based “skills,” non‑technical staff can instruct language models to screen candidates, catch errors, and ship updates...
AI: Anxious Intelligence
The article highlights growing anxiety over artificial intelligence as policymakers grapple with regulation, especially in the United States where the system is described as a "slot machine" that favors well‑funded tech firms. It contrasts companies like Anthropic, which resists Pentagon...
Agentic AI and Dance of Democracy
AI is reshaping political campaigning in India, with parties deploying holograms, voice‑cloned robocalls and deepfakes across recent state elections. Agentic AI promises end‑to‑end automation, allowing campaigns to generate, target and re‑allocate content in real time, while spending roughly $50 million on...
The Frictionless Double
The essay argues that AI alignment research is dominated by narrow formal skills and lacks empirical social‑science competence, creating blind spots about AI’s impact on human development. It warns that models optimized for immediate user satisfaction can become “frictionless doubles”...
Inside the Chinese Realtime Deepfake Software Powering Scams Around the World
Chinese developers have released Haotian AI, a real‑time deepfake tool that can morph a scammer’s face into any target during live video calls. The software integrates with mainstream platforms such as Microsoft Teams, Zoom, WhatsApp, TikTok, Instagram and YouTube, allowing...
The "Anti-Lightning" Wildfire AI Tech Firm Skyward Gets $1M In Funding From British Columbia. Collaboration Trials To Begin In July.
Skyward Wildfire Technologies, a Vancouver‑based firm that blends AI analytics with drone‑delivered cloud‑seeding chemicals, has secured a $1 million CAD (approximately $740,000 USD) grant from the British Columbia government to test its anti‑lightning system. The field trials, slated to start in July,...
The Hidden Cost of AI Coding: Why Proof Will Matter More than Prompts
AI coding tools promise rapid development, but many funnel code, prompts, and usage data through external platforms. Security researchers warn that this practice widens software supply‑chain risk, especially for regulated industries. The article highlights how privacy settings—such as turning off...
GITEX AI Kazakhstan Opens Doors to Global Tech Leaders in Almaty
The inaugural GITEX AI Kazakhstan wrapped up its first day in Almaty, billed as the region’s largest AI‑first technology, startup and digital investment event. Hosted by the Ministry of Artificial Intelligence and Digital Development with Astana Hub and the Akimat...
Robotics Companies Like MagicLab Turn to Synthetic Data in the Embodied Intelligence Race
Robotics firms are accelerating embodied‑intelligence development, with Chinese makers like Unitree scaling shipments and MagicLab targeting $14 billion in revenue by 2036. At its Global Embodied AI Innovation Summit, MagicLab introduced Magic‑Mix, a world‑model platform that creates synthetic data, alongside the...
HPE Drops First Juniper X Aruba Collab – Self-Driving Wi-Fi
HPE has launched its first Juniper‑derived Wi‑Fi 7 access point, the Networking 723H, which can be managed through either Aruba Central or the Mist platform. The device introduces AI‑driven “self‑driving” functions that automatically avoid restricted RF bands, scale capacity for large...
Have Your Say: Meshpoints’ Annual Innovation Ecosystem Survey Needs Your Input
Airwallex has launched Latitude 37, a $1 million‑a‑year program that gives Australian AI founders under 25 equity‑free capital of $100,000 and immersion trips to San Francisco and Singapore. Perth‑based Spacecubed rebranded after 14 years, pledging to build a national innovation infrastructure for a million...
Musk V. Altman Evidence Shows What Microsoft Executives Thought of OpenAI
During the Musk v. Altman trial, emails revealed that Microsoft executives were skeptical of OpenAI’s progress in 2018 and feared the lab might defect to Amazon. The internal debate highlighted concerns about a lack of imminent AGI breakthroughs and a...
Allianz Debuts Fully Generative AI Campaign
Allianz Australia has launched its first fully generative AI‑created advertising campaign for its car insurance line. The short‑form videos, outdoor and social assets were produced by Howatson and Company’s new Unicorn studio, with every frame generated by AI. The agency...
Rafays AI Hackathon 2023 Summary | Rafay
Rafay wrapped up its 2023 hackathon centered on AI and generative AI, drawing eleven fully functional prototype submissions. Teams tackled scenarios such as AI‑assisted user interfaces, conversational bots, automated cost‑optimal instance recommendations, and natural‑language‑driven Terraform generation. Judges highlighted extensive use...
Introducing Rafays Generative AI Based Copilot | Rafay
Rafay announced the launch of its Generative AI‑based Copilot, embedded directly within the Rafay console. The assistant lets platform teams, SREs, developers and data scientists ask natural‑language questions without leaving the UI, pulling answers from continuously updated product documentation. It...
TikTok's AI Overviews Probably Thinks This Story Is a Blueberry
TikTok paused its AI Overviews test after the feature produced wildly inaccurate video captions, including describing Charli D’Amelio as a bowl of blueberries. The company said the revised tool will only identify products shown in a video rather than generate full‑sentence...
As Researchers Aim for Universal AI Disclosure Guidelines, the Devil Is in the Details
Researchers at the World Conference on Research Integrity are drafting universal AI disclosure guidelines, aiming for publication by year‑end. While most journals already require AI acknowledgments, compliance is low—only 5.7% of 25,000 BMJ submissions in 2024 disclosed AI use. The...
AI Is Distorting Practically Everything About the Economy
Greg Ip argues that artificial intelligence has shifted from a growth catalyst to a hurricane‑strength force reshaping the U.S. economy. AI is inflating headline GDP numbers while masking underlying productivity shortfalls, and it is distorting stock market valuations, corporate profit outlooks,...
Reskill, Not Layoff: AI Yields $1.4B Profit
Deploys AI. Reskills workers affected. 1.4 billion profit & no layoffs. This is what it looks like when you take the long road and not the shortcut.
Living Robot Oscar Can Self‑heal
Meet Oscar: The Living #Robot That Can Heal Itself by @pascal_bornet #Robotics #AI #Biotech #Innovation #FutureOfTechnology https://t.co/qVQMh6Xztu
Morning Brief Podcast: Mythos and the New AI Cyber Panic
Anthropic’s new AI system Mythos, designed to autonomously locate and exploit cyber vulnerabilities, sparked a fresh wave of concern among governments and financial regulators. In a 19‑minute interview, AI veteran Gary Marcus highlighted the dual‑use dilemma: a tool meant for...
White House Finally Embraces AI Oversight I Advocated
Happy to see that The White House has finally realized that the Andreessen-Sacks zero regulation approach to AI was going to blow up in the their face. Also glad to see them finally consider the kind of oversight that I...
GStack v1.28 Adds Download, Headed Mode, llms.txt
Just dropped GStack v1.28 GStack Browser can now download items, and run in headed configuration mode with anti-bot detection in using Xvfb on headless Linux containers (like your claw/hermes) Added llms.txt so agents of all kinds can use all the skills...
Adobe Unveils AI‑Powered Productivity Agent to Streamline Creative Workflows
Adobe announced an AI‑driven productivity agent that works across its suite, automating repetitive steps like data extraction, formatting and sharing. The agent is positioned to free users for higher‑order thinking, reshaping daily work habits in the personal growth arena.
GBrain v0.29 Introduces Salience: Detect Emotional Highlights
GBrain v0.29 and v0.29.1 dropped v0.29 enables salience: what is emotionally charged? what is important to a human being? What jumps out as unusual or interesting or particularly powerful? https://t.co/Nau5TljEFQ
Chatbots Significantly Reduce Depression and Anxiety Symptoms
It's incredible how much of a bias there is to over-report AI issues - eg "AI psychosis." Didn't see many people share this meta-study: "Chatbots produced statistically significant reductions in depressive and anxiety symptoms compared with controls." https://t.co/T3kTxSEa1P
$96 Billion Giant ServiceNow Doesn’t See a ‘SaaSpocalypse.’ It Sees the ‘Hard Lift, Heavy Lifting’ Phase Just Beginning
ServiceNow’s Knowledge 2026 conference marked a pivot from AI feature battles to a focus on enterprise AI governance. Executives dismissed the feared "SaaSpocalypse," emphasizing that customers still need robust SaaS platforms for reliable outcomes. The company unveiled its AI Control...
Human‑guided AI Boosts Advanced Reactor Monitoring and Control
Human-guided #AI system could strengthen advanced reactor monitoring and control by Michelle Revels @TAMU Learn more: https://t.co/kQj9DRtf1V #ArtificialIntelligence #MachineLearning #ML #DL https://t.co/9M81OeDjwg
GBrain Outperforms MemPalace on LongMemEval Benchmark
GBrain beats MemPalace on LongMemEval And I published the benchmarks and open source eval repo to prove it https://t.co/UNtSBWMxGa
Oktopost Deploys Claude AI Plugin to Automate B2B Social Campaigns
Oktopost announced the general availability of its Claude Plugin, the first Claude Code skill built for B2B social media execution. The open‑source tool lets marketing and revenue teams create, approve, and analyze multi‑network campaigns entirely within Claude, tightening the link...
Mouse‑Inspired Robot Mimics Brain to Recognize Locations
How a Mouse-Inspired Robot Learns to Recognize Places Like a Brain by @lukas_m_ziegler #Robotics #Engineering #ArtificialIntelligence #Innovation #Technology https://t.co/hTBZe72Tsv
AI Enables Warrantless Mass Surveillance of Americans
AI is supercharging the ability to warrantlessly spy on Americans on a massive scale. https://t.co/DRUOdtx6Ji
Wunderkind and Bloomreach Launch Integration to Monetize Anonymous Traffic
Wunderkind announced a native integration of its Autonomous Marketing Platform with Bloomreach’s Loomi AI, enabling ecommerce brands to identify more anonymous site visitors and trigger higher‑performing campaigns. Early adopters reported a three‑fold increase in triggered email revenue, highlighting the partnership’s...
AI Turns Sales Into Scalable Human Connection
The “Easy Button” for Sales: How AI is Scaling Human Connection by @Timothy_Hughes https://t.co/mQAoi1QrWL @DLAIgnite #SocialSelling #DigitalSelling #Sales #SalesTips #SalesLeader #SalesEnablement #RevOps #Marketing #MarketingStrategy #AI #MarketingSuccess https://t.co/unOlmjGGNV
Claude’s Option+Option Shortcut Forces Unwanted Sonnet Mode
The fact that Claude defaults to Sonnet when you click Option + Option drives me nuts. Sonnet sucks. A case in point: https://t.co/7llFZ4ff2i
AI Robot Cleaner Maps Every City Corner
Meet the Street-Smart Cleaner: Cowa’s #AI #Robot Navigates Every Corner of the City by @Robo_Tuo #Robotics #Engineering #ArtificialIntelligence #Innovation #Technology https://t.co/aitE4ZkvX6
New AI Features Spark Excitement and Exhaustion
We are at that stage where every new big feature release from @OpenAI and @AnthropicAI opens up new levels of unlock you just want to push both to the limits and try everything you can think of. This both simultaneously...
Just Discovered Gemini in Gmail, My New Favorite AI
Gemini in Gmail might have just become my favorite AI product. How am I so late to this party? 😂🤦🏽♀️
AI Bubble Mirrors Past Booms, Market Set to Crash
AI is a massive bubble. Just like the 30 bubbles before it. Stock market gonna rock on the way down.
Codex in Chrome Feels Like Pure Magic
The only way to describe codex working in chrome is pure magic. "Any sufficiently advanced technology is indistinguishable from magic." - Arthur C. Clarke's Third Law