Today's AI Pulse
OpenAI launches ChatGPT Images 2.0 with multilingual text and real‑time web research
OpenAI has introduced ChatGPT Images 2.0, a next‑generation image model that adds multilingual typography, real‑time web research, and agentic reasoning to its generation pipeline. The new “Thinking” mode lets paid users produce up to eight coherent images per prompt, including floor plans, maps, UI mock‑ups, and apply capabilities to uploaded content.
Also developing:
Duolingo CEO Drops AI Metric From Reviews, Refocuses on Job Performance
Duolingo CEO Luis von Ahn announced that AI usage will be stripped from employee performance reviews, refocusing assessments on concrete job outcomes. The reversal follows a year of internal pushback after an April 2025 memo tied AI adoption to compensation and contractor reductions.
Google CEO Sundar Pichai Unveils Agentic AI Search and Urges Responsible AI Leadership
Google’s chief executive Sundar Pichai said the company is moving from traditional link‑based search to an "agentic" model that can book restaurants and handle complex tasks, while also urging America to lead AI development boldly and responsibly. The remarks, made...
Oracle Boosts Corporate Banking with Agentic AI Suite and Lucinity Investigation Tools
Oracle Financial Services announced the extension of its Agentic AI platform to corporate banking and the integration of Lucinity's AI‑led investigation tools. The moves embed generative AI agents in treasury, trade finance and lending workflows while giving banks a new...
Tata Power Chooses Databricks to Power AI‑Driven Energy Management
Tata Power, the Indian utility giant, has selected Databricks as its enterprise‑wide AI and data platform partner. The move will unify data engineering, analytics and AI across the company’s grid, renewable, and consumer‑facing businesses, accelerating real‑time insights and smarter energy...
AI Alliance Launches Project Tapestry to Build Open, Sovereign AI Platform
The AI Alliance, a nonprofit coalition of more than 200 members, announced Project Tapestry, an open-source platform for distributed training of frontier models. The initiative seeks to reconcile open AI development with sovereign data controls, offering participants the ability to...
Eclipse Ventures Raises $1.3 B to Build a Portfolio of Physical‑AI Startups
Eclipse Ventures closed a $1.3 billion fundraising round on April 7, 2026, splitting $720 million into an early‑stage Fund VI and $591 million into an Early Growth Fund III. The capital will be used to launch and scale startups that combine artificial intelligence with robotics, autonomous...
PIA Automation Launches Humanoid Robotics Business Segment, Partners with AGIBOT
PIA Automation announced a dedicated humanoid robotics business segment and a joint venture with AGIBOT called Joybot Manufacture. The move adds embodied AI‑enabled robots to its portfolio and includes plans for European manufacturing, signaling a new competitive thrust in industrial...
Demandbase Unveils AI-Powered Pipeline Engine to Boost B2B GTM Results
Demandbase introduced Demandbase AI at its GO London conference, positioning it as a unified AI pipeline engine for modern go‑to‑market teams. The platform leverages Context Intelligence, large‑language‑model integrations and a Site Customization Agent to turn noisy GTM signals into measurable...
OpenAI Launches GPT‑5.4‑Cyber, a Defensive AI Model for Cybersecurity
OpenAI introduced GPT‑5.4‑Cyber, a new defensive AI model aimed at bolstering cybersecurity defenses. The model mirrors Anthropic's Claude Mythos in its “cyber‑permissive” design and will be restricted to security‑focused deployments, prompting mixed reactions about its originality.
The Double-Edged Sword of AI in Crypto Trading
In 2026 AI has become a core engine of cryptocurrency trading, delivering millisecond‑level execution and 24/7 market surveillance that dwarf human capabilities. Platforms such as Token Metrics and BitsStrategy now give retail traders institutional‑grade analytics, enabling detection of micro‑price shifts,...
Strategic Small-Scale AI Projects Deliver Real ROI
Companies seeing real ROI from AI do one thing differently: They start small.. but strategic. Not “let’s try AI” But: “Let’s automate customer inquiries.” “Let’s improve response time.” Focused use cases > scattered experiments.
How to Thrive in the Age of AI
The article argues that AI literacy—being able to prompt, verify, and collaborate with AI—is now more critical than deep technical expertise. It cites Stanford’s AI Index showing 78% of firms used AI in 2024 and McKinsey’s finding that workflow redesign...
All‑in‑One Claude Desktop Unifies Parallel Development
One window. Multiple Claude sessions. Total control. The new Claude Code desktop experience lets you manage parallel workflows without losing context. Built-in terminal, file editing, HTML and PDF previews, and a faster diff viewer make it a complete workspace. And your CLI...
Vertiv's $15B Backlog Highlights Underpriced AI Cooling Play
Vertiv's $15B backlog proves AI's hunger for power & cooling. They're the liquid cooling leader—a critical "picks & shovels" play for data centers that Wall Street might still be underpricing. AI
Smart Catering: Reducing Cabin Food Waste with AI
Airbus and Virgin Atlantic have piloted an AI‑driven “Smart Catering” system to track in‑flight meal consumption and inventory. The technology uses camera‑based scanning on crew tablets to automatically log unused food and beverages, sending data to a ground cloud for...
On the Effects of Artificial General Intelligence on Transport
The article explores how Artificial General Intelligence (AGI) could transform transport by eventually automating all modes—from cars and trucks to trains, planes, and ships. In the near term, humanoid robots may act as interim operators, allowing legacy fleets to stay...
Bessent Calls Anthropic’s Mythos a Breakthrough in China AI Race
U.S. Treasury Secretary Scott Bessent praised Anthropic PBC’s new AI model Mythos as a breakthrough that will keep America ahead in the global AI race. He told a Wall Street Journal event that the United States remains three to six...
Anthropic Shifts Claude Enterprise to Usage-Based Billing, Potentially Doubling or Tripling Costs for Heavy Users
Anthropic has replaced its flat‑rate Claude Enterprise plan with a usage‑based model, charging a $20 per‑user monthly base plus compute fees. The change follows soaring demand for Claude Code and Claude Cowork, which have squeezed subscription margins. Analysts warn that...
AI - From Hype to Helpful
Dean, a self‑described non‑early‑adopter, outlines how he integrates generative AI—primarily ChatGPT and Claude—into his investment research and daily workflow. By templating conference briefs and company primers, he claims to shave 6–8 hours and over four hours respectively from routine tasks....
Pacvue Launches AI‑Powered Agent, Promising 200× Faster Media Workflows
Pacvue unveiled Pacvue Agent, an AI‑driven engine that integrates insight, recommendation and execution for commerce media. The company says early adopters are seeing workflow acceleration of up to 200 times, insight generation 80 times faster, and performance lifts of up...
Like Anthropic, OpenAI Will Share Latest Technology Only With Trusted Companies
OpenAI announced a limited rollout of GPT‑5.4‑Cyber, an AI model that scans software for security flaws. The initial phase will reach hundreds of trusted partners, with plans to expand to thousands in the coming weeks. The approach mirrors Anthropic’s recent...
UST Unveils Nimbus to Redesign Global Capability Centers with Advisory‑Led, Design‑Driven Model
UST announced Nimbus, a new advisory‑led, design‑driven offering that helps Global Capability Centers shift from cost‑saving units to strategic innovation engines. The service leverages UST’s engineering, AI and scaling expertise to deliver a build‑operate‑scale model that retains talent and intellectual...
California Bar Charges Two Lawyers Over Generative AI Misuse, Adds Third to Discipline Queue
The State Bar of California has filed disciplinary charges against attorneys Omid Emile Khalifeh and Steven Thomas Romeyn for submitting false court filings generated by generative AI, while a third lawyer, Sepideh Ardestani, faces a disciplinary stipulation. The actions underscore...
AWS Unveils Amazon Bio Discovery, Cutting Drug Design Time From Months to Weeks
Amazon Web Services launched Amazon Bio Discovery, an AI‑driven platform that lets researchers design and evaluate drug molecules without writing code. Early adopters such as Bayer, the Broad Institute and Memorial Sloan Kettering report that the tool can compress months‑long...
Lantern Rebrands, Expands AI‑Native Platform to Serve SMB HR and Operations
Lantern Partners AI announced today that it will operate under the new name Lantern, a move that signals a broader push into AI‑driven SaaS solutions for small‑ and medium‑size businesses. The rebrand highlights its LanternBRP™ platform, which unifies data across...
GeoComm Teams with Unknot.id to Deploy Public‑Safety‑Grade Indoor Location Intelligence
GeoComm and Unknot.id announced a partnership that embeds Unknot.id's Fuse AI™ platform into GeoComm's dynamic mapping suite, delivering public‑safety‑grade indoor location intelligence for government agencies and large enterprises. The joint solution aims to cut misrouting, speed response times, and improve...
Bairong Unveils Enterprise AI Platform with 200,000 Virtual Employees
Bairong Intelligence demonstrated a full‑scale AI workforce platform that assigns 200,000 virtual employees to real business roles. The launch, highlighted by a Harvard Business School delegation on March 30, marks a strategic pivot from AI as a tool to AI as...
Palo Alto Networks Finalizes $400 Million Koi Acquisition, Expands AI Endpoint Security
Palo Alto Networks has closed a $400 million deal to acquire Israeli cybersecurity startup Koi, bolstering its AI‑driven endpoint security portfolio. The acquisition integrates Koi’s control‑layer platform into Prisma AIRS and Cortex XDR, creating a new “Agentic Endpoint Security” category for CIOs planning...
Waymo Starts London Robotaxi Tests with 100 Jaguar I‑Pace Fleet
Waymo has begun testing its self‑driving Jaguar I‑Pace fleet on public roads in London, covering a 100‑square‑mile zone. The trial, run with human safety operators, is a prerequisite for the company’s planned 2026 commercial robotaxi service, pending UK regulatory approval.
Major Crypto Exchanges Including Coinbase and Binance Are Racing to Access Anthropic’s Mythos Model to Defend Against AI-Powered Attacks
Major crypto exchanges are scrambling to secure Anthropic’s new Mythos AI model to protect against AI‑driven attacks. Coinbase’s CSO confirmed close talks with Anthropic, while Binance is already testing Mythos alongside its own tools. Fireblocks reported that Claude Opus 4.6 exposed...
Earbuds with Tiny Cameras Enable Users to Chat with AI About Their Surroundings
A startup has unveiled earbuds that hide a sub‑millimeter camera, streaming live video to an on‑device AI assistant. Users can ask the AI to describe objects, read signs, or provide contextual information about their environment, all via voice. The device...
If AI Answers Everything, What Becomes of Us?
The piece argues that generative AI is moving beyond answering questions to pre‑filtering reality, delivering information before people experience it firsthand. By collapsing the gap between curiosity and answer, AI threatens the very tension that fuels imagination, learning and cultural...
Amplify Care Offers AI Cybersecurity Training
Amplify Care has launched an "AI and Cybersecurity" course within its Shield Training program, aimed at Canadian physicians navigating AI‑driven clinical systems. The offering combines expert‑led instruction with up to 12.5 Mainpro+ continuing education credits, addressing a sector where 64%...
Google DeepMind and Microsoft Research Unveil Agentic Risk Standard to Anchor AI in Finance
Researchers from Google DeepMind, Microsoft Research, Columbia University, t54 Labs and Virtuals Protocol introduced the Agentic Risk Standard (ARS), a settlement‑layer framework that imports financial risk‑management methods into AI agent transactions. The ARS aims to boost trust, accelerate fintech adoption...
Cloudera Launches Hybrid Data Platform with 2032 Support to Cut Costs and Speed AI
Cloudera announced a next‑generation hybrid data and AI platform that will receive support through 2032. The upgrade promises unified cloud‑on‑premise operations, reduced infrastructure spend and accelerated analytics for large enterprises.
Helical Secures $10 Million Seed Round to Launch Virtual AI Lab for Pharma
Helical Ltd., a biotech AI startup, closed a $10 million seed round led by Redalpine Venture Partners, with participation from Gradient, BoxGroup, Frst Capital and several AI CEOs. The funding will accelerate deployment of its virtual AI lab that turns foundation...
Adobe's 2026 AI Report Shows 80% Demand for Real‑Time CX Personalization
Adobe released its 2026 AI and Digital Trends report, revealing that 80% of surveyed executives want real‑time, highly personalized customer experiences. The study also flags data fragmentation and misaligned leadership as the chief obstacles to scaling AI across SaaS‑based CX...
Meta President Says Mark Zuckerberg Moved His Desk Into the AI Lab and Is Coding Alongside Alexandr Wang and Nat...
Meta President Dina Powell McCormick revealed at the World Economy Summit that Mark Zuckerberg has moved his desk into Meta's Superintelligence Labs, working side‑by‑side with lab lead Alexandr Wang and former GitHub CEO Nat Friedman. Zuckerberg is reportedly coding five...
AI Models Map Worm Brain, Scale to Synthetic Biology
The C. elegans nervous system has exactly 302 neurons and 7,000 synaptic connections. Scientists have had the full wiring diagram since 1986. What they haven't had: a computational model that captures how those neurons actually develop and wire themselves up. Eric Wasiolek...
Need Seamless Session Sync Between Claude Desktop and Mobile
I'm enjoying using the new Claude Code app (RIP terminal). How can I work on the same session across the Claude mobile app too? /remote-control doesn't seem to work and tbh I shouldn't have to type this to work across...
Microsoft Bets $10 Billion to Boost Japan's AI, Cybersecurity
Microsoft announced a $10 billion investment to expand AI infrastructure, cybersecurity services and local data centers in Japan, more than tripling its total spend in the country since 2024. The pledge includes partnerships with Sakura Internet and SoftBank and a commitment...
AI Lets One Vision Craft Beautifully Designed Software
AI makes software more tasteful By reducing the need for human coordination - between front-end devs, designers, PMs, etc. - a single opinionated person can make exquisitely designed software Long taste
Embracing AI: I Mandate Its Use in Class
Good advice. I *require* students to use AI in my class (I know many professors who ban it.)
Australia’s AI Moment: Building Asia–Pacific’s Compute Hub
McKinsey projects AI could generate up to AU$6.8 trillion (US$4.5 trillion) in global economic value, with Australia poised to capture a share by becoming an Asia‑Pacific compute hub. To meet projected AI‑driven demand, the country would need to expand data‑centre capacity from...
AI Referees Redefine Beach Volleyball: Enhancement or Soul‑Shift?
#AI Referees the Sand: Real-Time Beach Volleyball Analysis — Enhancing the Game or Changing Its Soul? by @measure_plan #ArtificialIntelligence #MachineLearning #ML https://t.co/kZRDd1LgfR
Compute Limits Trap AI: Stalling Growth Now and Later
Compute constraints are a double bind: On the inference side you need to either (a) raise prices, (b) ration use, and/or (c) serve worse models. This hurts current growth On the training side, you can't train the next gen of models to...
Printed MoS2 Memristive Nanosheet Networks for Spiking Neurons with Multi-Order Complexity
Researchers at Northwestern University have aerosol‑jet printed graphene/MoS₂/graphene memristive networks that exhibit snap‑back negative differential resistance and volatile threshold switching. The printed devices operate on flexible substrates, delivering oscillatory neuron circuits with tunable spiking frequencies up to 20 kHz and surviving...
Rippling AI Launch Sparks 78% YoY Revenue Surge
Rippling AI was the most successful launch we've ever done. On the heels of this launch, Rippling's revenue is now growing 78% YoY (at ARR over $1 Billion). And this growth rate has now increased, every quarter, for three straight...
What 27,000 AI Sessions Taught Us About How People Use Agents
Amplitude analyzed 27,410 sessions of its Global Agent, scoring each with 19 LLM‑as‑judge evaluators. The study found that 32% of successful sessions relied on users correcting the agent, while 97.7% of conversations showed no explicit thumbs‑up/down feedback. About a third...
Feature Selection Leads to Divergent Neurobiological Interpretations of Brain-Based Machine Learning Biomarkers
A new study in Nature Human Behaviour shows that the common practice of univariate feature selection in brain‑based machine‑learning models discards many informative connections. By partitioning connectome edges into ten non‑overlapping deciles, the researchers demonstrated that mid‑ranked and lower‑ranked feature...