Today's AI Pulse
OpenAI launches ChatGPT Images 2.0 with multilingual text and agentic reasoning
OpenAI has introduced ChatGPT Images 2.0, a next‑generation image model that adds multilingual typography, real‑time web research and agentic reasoning. The new “Thinking” mode lets paid users generate up to eight coherent images per prompt and create floor plans, maps and UI mock‑ups. The rollout expands OpenAI’s generative‑AI capabilities beyond text‑only outputs.
Also developing:
The Sequence Knowledge #846: Beyond Transformer: A New Series
The blog announces a new series tracking emerging alternatives to the Transformer architecture, which has dominated AI for a decade. It notes that self‑attention became popular because it mapped well onto GPUs and offered a simple token‑wise interaction model. The author argues that a “vibe shift” on arXiv signals growing interest in architectures that discard or extend self‑attention. The series will map concepts such as global text canvases, continuous control, and Liquid Foundation Models.
Here's Why the AI Infrastructure Story Is Just Getting Bigger for GOOGL
Alphabet is expanding its AI‑chip supply chain by partnering with Marvell Technology to co‑develop a memory‑processing unit and an inference‑focused TPU, complementing its existing ties with Broadcom and MediaTek. The company has pledged $175‑$185 billion in capital expenditures for 2026, underpinned...
IQIYI Targets AI-Generated Majority of New Films and TV in Five Years
iQIYI announced that AI will produce the bulk of its new films and television shows within five years, unveiling a slate of 16 AI-generated sci‑fi and anime movies and a summer‑targeted fully AI‑made feature. The move signals a strategic shift...
What AI's Impact on Engineering Tells Us About Where Org Design Is Headed
AI tools are accelerating software development, leading to a surge in senior engineering job postings despite earlier predictions of engineer displacement. US developer openings have risen for six months, reaching three‑year highs, while entry‑level hiring has dropped 25% year‑over‑year. Companies...
IFS Policy Brief Explores the Debate Over AI and Personhood
The Institute for Family Studies policy brief argues that the growing push to grant legal personhood to artificial intelligence is rooted in analogies to corporate and animal‑rights law. It warns that extending personhood could insulate AI firms from liability, amplify...
Anthropic Overhauls Leadership as IPO Looms, Targeting $630B Valuation
Anthropic announced a sweeping leadership restructuring, promoting co-founder Mike Krieger to co‑lead its new Anthropic Labs and appointing Ami Vora as chief product officer, as the AI startup readies for a potential October 2026 IPO that could fetch a $630 billion...
Coforge Rolls Out Voyager.AI and Flightflex.AI to Transform Airline Operations
Coforge announced the launch of Voyager.AI and Flightflex.AI, two AI‑native platforms designed to boost airline revenue and passenger experience. The tools deliver real‑time personalized offers and automated disruption management, positioning Coforge as a strategic partner in airline digital transformation.
Cadence and Nvidia Expand AI Partnership to Speed Digital‑Twin Workflows for Real‑Estate Engineering
Cadence Design Systems and Nvidia announced an expanded partnership that integrates Nvidia’s accelerated computing and Omniverse platform into Cadence’s EDA tools. The collaboration promises speed improvements of up to 100 times and productivity gains of up to 10 times, aiming...
AltaClaro Launches DepoSim, AI‑Driven Deposition Simulator for Lawyers
AltaClaro introduced DepoSim, an AI‑driven deposition simulator built with Verbit, allowing lawyers to practice live oral depositions on demand. The platform promises repeatable, scalable training and instant rubric‑based feedback, addressing long‑standing cost and consistency issues in litigation preparation.
SUSE and Nvidia Unveil SUSE AI Factory, a Sovereign Enterprise AI Platform
At SUSECON 2026 in Prague, SUSE and Nvidia introduced SUSE AI Factory, a pre‑validated, turnkey AI platform designed for enterprises and governments that need digital sovereignty and strict security. The solution integrates SUSE Rancher Prime, SLES and Nvidia AI Enterprise...
GitLab 18.11 Launches AI‑Driven SAST Agent and Automated Merge‑Request Generation
GitLab released version 18.11, adding a platform‑native AI SAST remediation agent that auto‑generates merge requests and two new AI assistants for CI pipeline design and real‑time data analysis. The features aim to close the “AI paradox” by extending AI beyond...
Cognizant Unveils Skillspring AI Platform to Upskill Enterprise Workforce
Cognizant announced the launch of Skillspring™, an AI‑native, multimodal learning platform that maps skills to roles and projects. The service targets the $4.5 trillion of U.S. work tasks now handled by AI and the 93% of jobs already affected, promising faster...
ServiceNow and Tesla Set to Report Earnings Amid AI Growth and Inventory Concerns
ServiceNow and Tesla are slated to release their first‑quarter earnings after the market closes on Wednesday, April 22. ServiceNow will showcase AI‑driven subscription growth, while Tesla faces a 12% inventory buildup and seeks to highlight its expanding energy business.
OpenGov Names Thiago Sá Freire CEO to Accelerate AI‑Native Growth
OpenGov announced that its president and COO, Thiago Sá Freire, will assume the chief executive role on April 1, 2026. The move follows Cox Enterprises' 2024 majority‑ownership acquisition and signals a push to deepen AI‑driven services for more than 2,000 U.S. government...
On Taiwan’s Breathtaking Order Boom
Taiwan’s Ministry of Economic Affairs reported a near‑70% year‑over‑year jump in AI‑hardware export orders for March, the strongest surge since a 72% rise in January 2010. Economists had only expected a 45% increase, underscoring the unexpected strength of global demand. Orders...
AI Hiring Boom: How Cognitive Technologies Are Redefining E-Commerce Recruitment
E‑commerce retailers are moving from isolated AI tools to unified cognitive AI platforms that combine natural language understanding, vision, knowledge graphs, and real‑time inference. These platforms power conversational interfaces that can remember context, ask clarifying questions, and execute actions such...
Fraudsters May Target AI Mandates as Agentic Commerce Takes Off
AI‑driven agents are set to handle consumer transactions, a trend dubbed agentic commerce. McKinsey projects up to $1 trillion in U.S. B2C revenue by 2030, with a global market potential of $3‑5 trillion. Visa’s security unit observed a 450% surge in dark‑web...
M42 Study Leverages 500,000 Genomes to Spot Vision‑Loss Risks in UAE
M42 announced that analysis of more than 500,000 Emirati genomes uncovered around 100 genetic drivers of inherited eye disease. The partnership with Abu Dhabi's Department of Health demonstrates how massive genomic datasets and AI can shift eye care from treatment...
Cognizant Propels AI Workforce Training with Cognizant Skillspring™: New Talent Transformation Platform Designed to Accelerate Clients’ Workforce AI Readiness
Cognizant unveiled Skillspring™, an AI‑native, multimodal learning platform that embeds conversational tutoring directly into daily workflows. The solution links skill development to specific roles, projects and performance outcomes, and includes an AI Fluency Dashboard that tracks individual readiness in real...
ANALYSIS: Big Tech Sets AI to Catch AI
Advanced AI is reshaping cyber‑security as both a weapon and a shield. Hackers leveraged over 1,000 AI prompts to breach Mexico’s tax authority, exposing 195 million records and prompting one of the largest government data leaks. At the same time, Anthropic’s...
AI Is The New Ad Space
Debra Aho Williamson, founder of Sonata Insights, will keynote AdExchanger’s Programmatic AI conference in Las Vegas (May 18‑20). She argues that AI tools like ChatGPT, Copilot, and retail assistants constitute a new media environment—“AI media”—that can guide consumers from query to...
BlackBerry QNX Deepens Nvidia Edge AI Tie‑up, Fuels Enterprise Growth
BlackBerry's QNX division announced an expanded integration of its Safety 8.0 OS with Nvidia's IGX Thor edge AI platform at Hannover Messe, while Chinese EV maker Leapmotor selected QNX for its new D19 SUV. The moves broaden QNX's reach beyond automotive...
Cerebras Systems Files Nasdaq IPO, Adding a Public Contender to AI Chip Market
Cerebras Systems, the Silicon Valley AI chipmaker behind wafer‑scale processors, filed a registration statement with the SEC to list on Nasdaq under the ticker CBRS. The move follows a 2025 revenue surge to $510 million and a $20 billion investment from OpenAI,...
The AI “Sandwich”: Why HR Teams Are Caught in the Middle
In this episode Tim Fisher talks with Josh Rod of HiBob about how AI is reshaping HR. They explain HiBob’s all‑in‑one HCM platform and why AI is a strategic priority for delivering real value rather than hype. The conversation highlights...
How Using AI Skills for Law Firm Workflows Can Turbocharge Your SOPs
Law firms are replacing static standard operating procedures with custom AI Skills built inside Claude, an advanced language model. These Skills are plain‑text markdown files that encode drafting rules, intake steps, and billing preferences, allowing the AI to execute tasks...
Homeland Security Reportedly Wants to Develop Smart Glasses for ICE
The Department of Homeland Security is developing "ICE Glasses," smart eyewear that captures video and runs facial‑recognition and gait‑analysis algorithms to identify individuals in real time. Budget documents show the prototypes aim for field deployment by September 2027, giving ICE...
How Rheinische Post Built an AI Governance that Aligns with Its Digital Subscriber Goals
Rheinische Post Mediengruppe (RPM) built an AI governance model that directly supports its goal of 250,000 digital subscribers. Under Group Head of AI Margret Seeger, the publisher aligned AI activities with its broader digital transformation, creating a three‑pillar framework (Head...
Your AI Can’t Read an Invoice. That Should Worry You More than Whether It Can Pass a Math Exam
Enterprise software veterans note that large language models can solve Olympiad‑level math yet still miss basic totals on invoices. The author argues this gap isn’t due to reasoning limits but to perception challenges—messy layouts, poor scans, and ambiguous formatting. While...
AI Enhances SEO, Yet Human Expertise Remains Crucial
I created an AI tool to improve internal linking and enhance meta titles and meta descriptions. After a few weeks, I can tell it's working nicely (we started where you see the dip in the graph around mid-march) But you have to...
Genius Sports Inks Multi-Year Agreement with Swiss Football League
Genius Sports signed a multi‑year deal with the Swiss Football League to supply its GeniusIQ data and AI platform across the Super League and Challenge League starting in the 2026/27 season. The agreement will install a unified intelligence layer covering...
Simple CLAUDE.md File Boosts AI Output with Karpathy's Principles
INSANE. Did you know the #1 GitHub trending repo (68k+ stars) is literally just a CLAUDE.md file 🤯 It uses @karpathy's 4 LLM principles to keep AI in check: → Seek clarity: Always ask before making assumptions. → Stay minimal: Avoid bloat and keep...
The Global AI Threat Has Arrived
Anthropic unveiled Claude Mythos Preview, an AI model that can autonomously locate and exploit vulnerabilities in major operating systems and web browsers. The discovery has alarmed business leaders and policymakers worldwide, prompting concerns about a new class of AI‑driven cyber threats. Even...
AI Could Help NHS Clinicians Diagnose Childhood Sleep Apnoea, Study Finds
Seluna, a Glasgow‑based health‑tech firm, piloted its cloud‑based AI autoscoring software on 500 retrospective paediatric sleep studies from NHS Greater Glasgow and Clyde. The algorithm flagged severe sleep apnoea with 100% accuracy, while mild and moderate cases were identified with...
Why Are Respected Film-Makers Suddenly Embracing AI?
Steven Soderbergh’s new film *The Christophers* spotlights his renewed interest in generative AI, which he plans to use for dream‑like imagery in a forthcoming John Lennon documentary and a Spanish‑American War feature. His ambivalent comments have sparked debate as other...
AI Generates Code, Developers Clean Up the Mess
Vibe coding was supposed to kill software developers. What strikes me is how quickly that story flipped. We were told AI would replace developers. Instead, it seems to be creating a new role: 𝗩𝗶𝗯𝗲 𝗖𝗼𝗱𝗶𝗻𝗴 𝗖𝗹𝗲𝗮𝗻𝘂𝗽 𝗦𝗽𝗲𝗰𝗶𝗮𝗹𝗶𝘀𝘁™. And honestly, that feels very plausible. Because when...
Court’s Ruling Offers Cautionary Tale for Clients Using Generative AI
In February 2026, Judge Jed S. Rakoff ruled that a client’s communications with the Claude generative‑AI platform were not shielded by attorney‑client privilege or the work‑product doctrine. The decision hinged on the lack of a lawyer in the exchange, Claude’s...
How to Prepare Young Students for an AI World with Dr. Nneka McGee – Easy EdTech Podcast 367
Podcast host Dr. Monica Burns talks with AI‑education strategist Dr. Nneka McGee about preparing young learners for an AI‑driven world using developmentally appropriate, unplugged activities. The conversation highlights conversation‑based techniques, pattern‑recognition games, and other hands‑on exercises that teach AI fundamentals without...
4 Higher Education Leaders on AI’s Biggest Benefits and Risks
At the ASU+GSV Summit, presidents from Radford University, Southern New Hampshire University, the University of North Texas and UC San Diego discussed AI’s transformative potential for higher education. They highlighted opportunities such as boosting employability for under‑prepared students, delivering personalized...
Instrumental Cooperation, Not Moral Alignment, Drives Human and Machine Values
I was having a great discussion with an X user about AI alignment. His thoughtful claim was that while it's easy to see machines instrumentally (based on logical needs) aligning with humans for their own survival, it's much tougher to...
German Central Bank Chief Advocates Widespread Access to Anthropic's Mythos
JUST IN: The German central bank chief is pushing for broad availability of Anthropic's Mythos.
Top AI Education Use Cases Share Core Principles
What The Best #AI Use Cases In Education Have In Common by @TalFrankfurt @Forbes Learn more: https://t.co/cHZnSGNnVN #ArtificialIntelligence #MachineLearning #ML https://t.co/zLJc3AtsWH
AI Speeds Coding, but Teams Prioritize More than Speed
AI writes code faster than ever — but smart teams know speed isn't the whole story. https://t.co/9DQh4r2Pwb
Humanoid Robots Transform Shopping in Hangzhou
The Future of Shopping: Humanoid #Robots Take Over Store Operations in Hangzhou by @CyberRobooo #Engineering #ArtificialIntelligence #Innovation #Technology https://t.co/qP9ZmkWQNw
Master Hermes Agent in 2‑Hour Course
.@JulianGoldieSEO just dropped a full 2-HOUR COURSE on @nousresearch's Hermes agent 🤯 Skip the Netflix binge and learn to build and automate ANYTHING with the agent everyone is raving about. https://t.co/tGwZHv3o76
Use Claude's Find Skills for Instant Perfect Suggestions
a Japanese dev literally found the ultimate Claude Code cheat code before the rest of us 💥💥💥 He simply installs 'Find Skills' and asks: 'Are there any good skills for [GOAL]?' It instantly hands you the perfect match out of 100s 👀 install guide...
Vendor Lock‑in Poses Greater Risk than Single‑chip Reliance
Lines between infrastructure and applications are blurring. Recognize the risks, rewards, and trade-offs of vendor lock-in. The bigger risk isn't NVIDIA providing chips and AI, but getting too tied to one provider. #TechTrends #VendorLockIn https://t.co/PE7LQp3hzn
New Plugin Tracks AI Prompt Water Use, Enables Offsetting
A new plugin enables you to see exactly how much water your AI prompts are using up, and make a donation to offset it by giving water back to communities in need. https://t.co/7PGqr71nj4
AI Tools Offer Immediate Opportunities, Education Lags Behind
Just finished 'The Infinity Machine'. Hassabis is right on career advice: "Immerse yourself in AI tools. Everyone has access to tools 3-6 months behind frontier. Enormous opportunity lies in applying AI to unexplored areas." Shame that many in education say...
Run a Private AI Stack Locally—No Cloud, No Fees
Want the perfect local AI stack? LOOK NO FURTHER 👀 OpenClaw + Gemma 4 + Ollama 🔥 Run a powerful local agent right on your own machine. → Zero cloud → Zero subscriptions → TOTAL privacy Check that rad demo from @JulianGoldieSEO ↓ https://t.co/Epthgad7Jk
AI Is a Powered Mimic, Not Self-Sustaining Life
An AI, regardless of its "behavioral and verbal sophistication, remains a functional mimic. Its existence is dependent on external energy and human design, lacking the intrinsic, self-sustaining processes characteristic of biological life" 💡 @AnnaCiaunica https://t.co/XZeEohCpkV