Today's AI Pulse
OpenAI unveils ChatGPT Images 2.0 with multilingual text and ‘Thinking’ mode
OpenAI rolled out 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, including floor plans, maps and UI mock‑ups. The feature expands the model’s ability to work with uploaded content.
Also developing:
A Photovoltaic Power Forecasting Method Integrating Physical Mechanisms and Deep Learning
Researchers have introduced a hybrid photovoltaic power forecasting method that merges physical modeling with deep learning. The approach uses a non‑uniform error compensation strategy, a 37‑dimensional feature system, and a dynamic weighted fusion based on four confidence factors. Validation on 15‑minute data from a city‑scale PV plant shows MAE of 3.571 kW, RMSE of 6.384 kW, and an R² of 0.9781, outperforming pure physical and pure neural models. The study highlights the value of adaptive hybrid techniques for grid‑level solar integration.
Hotels and Airlines Vs. NerdWallet and Reddit: Who’s Winning in AI Visibility
AI travel assistants are increasingly pulling information from third‑party sites rather than hotel or airline homepages. Limy’s analysis shows NerdWallet accounts for 13.6% of citations, outpacing Hyatt’s own site at 10.3%, while Reddit also ranks among the top sources. Direct...
Sam Altman's "Proof of Human" Company Pushes Into Mainstream Services
World, the identity‑verification startup co‑founded by Sam Altman, announced a wave of new integrations with Zoom, DocuSign, Tinder, Okta, Shopify and VanEck. The company upgraded its World ID protocol, open‑sourced it for any app, and launched a standalone app for...
A Hybrid Generative and Transformer-Based Framework for Anomaly Detection in Industrial Sensor Time-Series for Predictive Maintenance
The paper presents a hybrid anomaly‑detection framework that combines a state‑of‑the‑art GAN, statistical distance metrics and the Anomaly Transformer to monitor multivariate industrial sensor streams. By leveraging reconstruction error, Mahalanobis distance and isolation‑tree scores, the system captures both distributional and...
SAC-YOLO: Efficient Multi-Scale Feature Fusion for Transmission Line Defect Detection
The paper presents SAC‑YOLO, an enhanced YOLOv11n model tailored for transmission‑line defect detection. It swaps the SPPF module for an AIFI block to boost intra‑scale interaction, adds a C3K2‑CFBlock for efficient local‑global feature blending, and introduces a Semantic‑Guided Multi‑Scale Fusion...
Canva Becomes the Design Layer Inside Claude with New Anthropic Partnership
Canva and Anthropic have deepened their two‑year partnership by launching Claude Design, a feature that leverages Canva’s Design Engine to turn text prompts into fully editable, on‑brand visuals via Claude Opus 4.7. The product is part of Canva AI 2.0, which adds...
AI Divide: Exec Praise vs Employee Anxiety
"Different realities are happening inside the same company. Executives are celebrating time saved and progress made. employees are quietly paying an AI tax through rework, uncertainty, and rising anxiety. When those realities drift apart, AI does not scale." https://briansolis.substack.com/p/the-ai-divide-is-a-leadership-crisis
Oracle Delivers Semantic Search without LLMs
Oracle introduced Trusted Answer Search, a semantic search solution that relies on vector similarity rather than large language models. Enterprises define a curated search space of approved documents and metadata, enabling deterministic, auditable responses such as reports or URLs. The...
Most Enterprises Can't Stop Stage-Three AI Agent Threats, VentureBeat Survey Finds
A VentureBeat three‑wave survey of 108 enterprises reveals that most organizations rely on monitoring AI agents without enforcing controls or isolating workloads, a structural gap that leaves them vulnerable to stage‑three threats. While 88% reported AI‑agent security incidents in the...
NAB Show: FOR-A America Announces New AI Functionality On FOR-A IMPULSE and Immediate Delivery of MixBoard & HVS-Q12
At NAB Show 2026, FOR‑A America unveiled AI‑enhanced capabilities for its software‑defined IMPULSE live‑production platform, adding an AI‑assisted Graph Editor that auto‑creates node graphs and system diagrams. The company also announced the immediate availability of MixBoard, a GPU‑accelerated software switcher...
Vox AI Turns Drive-Thru Conversations Into Operational Intelligence
Vox AI, a conversational voice‑AI platform for drive‑thru quick‑service restaurants, launched Vox AI Insights, an analytics solution that converts every drive‑thru conversation into operational intelligence. The system hooks into existing audio equipment, captures full conversation context in real time, and...
The Wall Around Claude 4.7 Does Not Extend to Dread
Anthropic unveiled Claude Opus 4.7 on April 16 2026, intentionally reducing its offensive cyber capabilities while adding automated safeguards and a Cyber Verification Program for vetted defenders. Simultaneously, underground forums on Dread, Reddit, and Telegram circulated jailbreaks and a cross‑vendor prompt‑injection attack called...
Cold Circuits, Warm Heart: AI and Empathy in CX
Ian Jacobs argues that AI itself cannot feel empathy, but its responses can be perceived as empathetic, which is what matters in customer experience. He notes that AI can consistently recognize distress signals, use context‑aware language, and avoid human fatigue,...
How Patients Are Using AI to Find Healthcare Providers in 2026
Patients are now starting their care journey by asking AI models, not search engines. Recent surveys show roughly one‑quarter of U.S. adults used a health‑focused chatbot in the past 30 days, and 14% skipped a doctor visit after AI advice—about...
Solo Open-Source Projects Address Challenges of Agentic AI
Solo.io unveiled two open‑source projects, Agent Registry and Agent Evals, to tackle enterprise hurdles in adopting agentic AI. Agent Registry, now a CNCF sandbox, offers a centralized hub for governing AI agents, skills, and MCP tools on Kubernetes. Agent Evals...
How AI Is Reshaping Product Discovery in F&B
AI is creating a new "third shelf" where intelligent agents surface food‑and‑beverage products directly to consumers, bypassing traditional web searches. Brands are seeing a 30‑fold surge in ChatGPT‑driven traffic, signaling that shoppers are using generative AI as a purchasing tool....
FAA Quietly Developing AI-Enabled Predictive Air Traffic Management System
The Federal Aviation Administration is quietly building an AI‑driven tool called Strategic Management of Airspace Routing Trajectories (SMART) to predict and resolve air‑traffic bottlenecks before flights depart. Administrator Bryan Bedford is championing the effort, with Palantir, Thales and Airspace Intelligence...
Use YOLO, Not Opus, for Fast Accurate Detection
Don't use Opus 4.7 for computer vision. It's the wrong tool. I ran a simple pointing task on both Opus 4.7 and GPT 5.4: find the cars in an aerial image. Both took several minutes. That alone should...
Chinese Pharma Leads Speed, Insilico Wins on Novelty
Very insightful article in China Daily. There is a big delta between the AI CRO model and AI Biotech model. At Insilico, we do not compete with Xtalpi or other local companies. The most hardcore competitors now are the large...
The Great Beheading Begins
Salesforce unveiled "Headless 360," an architectural overhaul that turns every feature of its 27‑year‑old platform into APIs, command‑line tools, or micro‑service endpoints. The move enables autonomous AI agents to execute tasks across the CRM without ever opening a browser. Industry...
Brands Bought All The AI, Forgot To Fix The Chaos, Merkle Says
Merkle’s latest research reveals that while brands have splurged on AI tools, most still wrestle with tangled content workflows. Only 5% of companies report a clear ROI from internal AI projects, and 64% of executives cite competing initiatives as a...
25 Claude Skills that Give Your Startup a Marketing Team It Cannot Afford Yet
Early‑stage startups often stall because each marketing campaign requires rebuilding workflows from scratch, draining limited resources. The post proposes "Claude Skills"—reusable markdown prompt files that encode brand voice, positioning, and process logic, allowing a single hire to produce output equivalent...
14 Things Everyone Missed About Opus 4.7 Yesterday
Anthropic unveiled Claude Opus 4.7, a model that dramatically increases literal adherence to prompts, causing older prompt formulations to behave unpredictably. The upgrade retires Opus 4.6, meaning users must rely on archived outputs for any direct comparison. Anthropic also introduced...
The $600 Billion Home Services Market Runs on Clipboards (& How A.I Can Change That)
The U.S. home‑service sector comprises roughly 800,000 firms that generate about $600 billion in annual revenue, yet most still rely on paper tickets and Excel spreadsheets. This antiquated workflow costs companies 30‑40% of inbound leads and $50,000‑$80,000 each year in scheduling...
Experts Slam Non‑specialist Analogies in AI Debate
I love it when podcast hosts and newsletter compilers cite other non industry or non expert sources on a complex issue to try and dispute one of the architects of the entire AI revolution... Jensen on nuclear weapons analogy: "Comparing...
Integrate AI, Triple Margins; Pure AI Startups Fail
90% of the 'AI wrapper' startups are going to die in 12 months. but the founders who integrate AI into their existing businesses are going to 3x their margins. i'm not building an AI company. i'm building a company that happens...
AI’s Are Beginning To Get Emotional Intelligence
AI developers are adding emotional intelligence to chatbots and real‑time coaching tools, with startups like Amotions AI offering video‑call emotion analysis and major firms such as OpenAI, Anthropic, Google, and xAI touting warmer, more conversational models. These systems claim to...
Authentic Thought Beats Futurist Lies
The Future of Everything is Lies, I Guess https://aphyr.com/posts/420-the-future-of-everything-is-lies-i-guess-where-do-we-go-from-here "Refuse to insult your readers: think your own thoughts and write your own words." - Gonna be quoting this for a long while
OpenAI Does the Boring Sensible Thing
OpenAI unveiled GPT‑Rosalind, a specialized reasoning model for biology, drug discovery and translational medicine, and expanded its Codex platform into a general‑purpose workflow assistant. The company made the new model available via ChatGPT, Codex and API to qualified customers, naming...
MPN Video Podcast: Why Vertical Video Could Be Bigger Than Traditional TV for Sports
In a Media Play News podcast, Grabyo CEO Gareth Capon discusses the surge of AI‑powered vertical video for live sports. The platform enables broadcasters to reformat traditional feeds into mobile‑first streams, add real‑time clips, and localize content on the fly....
AI Mode Now Orders Nearby Items via Voice
Expanding from Search to AI Mode -> AI Mode can now call nearby stores for last-minute shopping "This tool launched directly on Search last November, and it’s rolling out in the coming weeks on AI Mode in the U.S." "Just describe what...
Agentic Web Arrives: 5 Key AI Search Shifts
The Agentic Web is Here: 5 AI Shifts I'm Discussing with Clients about AI Search - Marie Haynes https://buff.ly/VRW47KT
If the Shoe Doesn't Fit
Allbirds, the eco‑friendly shoe maker once valued at $4 billion, agreed to sell its brand for $39 million and rebrand the remaining assets as NewBird AI. The fledgling venture plans to raise $50 million from an undisclosed investor to purchase high‑performance GPUs and...
Tesla FSD Spots Road Debris Before Driver, Intervenes
I wasn't sure what this was (it was rather large) but Tesla's FSD saw the debris in the road sooner than I did and took aggressive action to ensure I didn't run it over.
AI Threatens Tech Moats, Future Cashflow Valuations Collapse
If AI lives up to the hype, it will erode the traditional long term moats of technology companies. As a result, no one with a brain wants to pay for cashflows far in the future. These multiples cannot come back unless...
AI Virtual Assistant Texted 1.1M Wisconsin Voters, Logging 21K Campaign Conversations
Defend Our Courts and startup Convos rolled out an AI‑powered virtual assistant that texted more than 1.1 million Wisconsin voters in the April 7 Supreme Court race. The bot recorded 21,316 substantive conversations while the group spent roughly $84,000, highlighting a...
Demanding 1,200 Benchmark Test Before Claiming SOTA
Until they test it in @InSilicoMeds benchmarks (1,200 benchmarks), I will not believe that this is anywhere close to SOTA. Let's benchmark it folks. Whoever has access and would like to benchmark - let's collaborate
Open‑weight AI Keeps Competition Alive, Prevents Monopoly
It's incredibly hard to stop open weight AI developers from distilling knowledge out of frontier models. Another reason to believe AI will remain hyper-competitive and not monopolized. Economics from @dwarkesh_sp newsletter this morning. https://t.co/KKocBrLgNs https://t.co/swXubgHvbM
Chef Robotics Escaped the Robot Cooking Graveyard and Says It’s Thriving — Here’s Why
Chef Robotics announced it has processed 100 million robot‑deposited servings, marking a key milestone in its pivot from fast‑casual restaurant automation to large‑scale food manufacturing. The company now serves enterprise clients such as Amy’s Kitchen, Chef Bombay, and a major school‑lunch...
Upload Updated Contract, AI Amends It Instantly
Renegotiated a contract? In JustPaid, upload the updated version to amend it in seconds AI handles the rest. https://t.co/Fl423xNbMo
WOZCODE Slashes Claude Code Costs by 54%
Someone built a plugin that cuts Claude Code's cost by 54% on the same task, same model. Not a different model. Not a different harness. Just smarter tools underneath. It's called WOZCODE and the benchmark numbers are genuinely CRAZYYYY. https://t.co/OUoCq9J5ep
BearingPoint Study Finds Over One-Third of Firms Lag on Climate Goals Amid AI Surge
BearingPoint’s latest thought‑leadership report shows that over one‑third of surveyed enterprises are already missing science‑based climate targets, citing AI’s growing energy demand as a key obstacle. The findings highlight a widening gap between sustainability pledges and operational execution.
AI Risk Lies in Agents' Actions, Not Model Statements
The New #AI Risk Isn’t What Models Say. It’s What Agents Do by Tyler Akidau @Forbes Learn more: https://t.co/nVK1YaBgb4 #ArtificialIntelligence #MachineLearning #ML #DL https://t.co/Fp5k8v0NUM
AI Agents Gain Business Autonomy Through Web3 Integration
Web 3 and AI mix. Meet the guys who are bringing your AI Agents the ability to run their own business: @eigencloud Conversation with @sreeramkannan founder about the shifts happening as Web 3 and the AI worlds collide. https://t.co/Q0SFBhdjp7
Influencers Are Replacing Themselves With AI Clones
Influencers are increasingly deploying AI-generated digital twins to extend their brand presence around the clock, prompting talent agencies such as CAA to treat clone management as a strategic priority. Legal counsel warns that without explicit consent, these likenesses could expose...
AI Generates Claude Design Video From Blog and Tweets
Made this 30 second video of Claude Design just by pasting in the Claude Design blog post and some tweets from @AnthropicAI employees Kinda speechless. https://t.co/9uFxO3LKzf
HeyGen Now Powers All My AI Agents
HeyGen just put itself inside every agent I use. This is awesome. Check it out:
AI Agents Will Flood Networks—New Strategies Needed
Given how much more Internet traffic AI Agents are going to create, we need new strategies to ensure the network can keep up. Here’s one…
Epic's AI Suite Crowds Out Third‑Party Tools in Healthcare
My new Substack: Epic, the leading EHR vendor, has released >100 AI tools. Choosing them over third-party tools has become the safe choice for many health systems. I explore why, and the bigger questions: can – and should – something be done about...
AI Trust Grows When Societies Become More AI‑Savvy
Societies that are smarter about AI tend to trust it more We should get better at educating people h/t @EricNewcomer https://t.co/MJpe1VWlAe