Today's AI Pulse
Odyssey lands $310M Series B, valued at $1.45B
Odyssey announced a $310 million Series B round that values the AI lab at $1.45 billion. The round was led by Natural Capital and included backers such as Amazon, AMD Ventures, GV, EQT and IQT. In parallel, Odyssey struck a preferred‑cloud deal with AWS to use Trainium silicon for large‑scale model training.
Also developing:
GitLab 19.0 Bets that the Real Bottleneck in Software Delivery Is Everything After Writing the Code
GitLab unveiled version 19.0, centering on "intelligent orchestration" that extends its Duo Agent Platform across the entire software lifecycle. The release introduces a generally available SBOM‑based dependency scanner for Maven, Gradle and Python, exposing hidden vulnerabilities in transitive libraries. It also adds support for Claude Opus 4.7, Google Gemini, and several open‑source models, while enhancing AI‑driven developer workflows and group‑level review policies. To manage AI costs, GitLab rolls out a credit‑based metering system for premium and ultimate tiers.
Google Explains How It Will Infuse Ads Into AI Answers
Google announced new AI‑driven ad formats as part of its Gemini 3.5 Flash upgrade to Search. The company will embed Conversational Discovery ads and Highlighted Answers within AI Mode responses while keeping traditional SERPs as the default. Advertisers will use...
From Automation to Execution: How Agentic AI Is Redefining the Role of Finance Teams
Sage is advancing finance from simple automation to agentic AI that can execute tasks within workflows, shifting finance teams toward exception handling and strategic oversight. The company emphasizes built‑in governance—transparent reasoning, human approvals, and audit trails—to ensure CFOs can trust...
Canada Injects Millions Into AI Research Program Amid “Global War” For Talent
The Canadian government announced a CAD 24 million (≈US $18 million) investment in the CIFAR AI Chairs program, adding 20 new chair appointments and renewing 22 existing ones. The funding supports 143 leading AI researchers across the country’s three premier institutes—Amii, Mila, and the...
Shopify Shareholder Proposal Pushes for Responsible AI Policy Ahead of June 16 Annual Meeting, Urged to Vote Against by Board
Shopify faces a shareholder proposal from the Shareholder Association for Research and Education (SHARE) urging the company to adopt a responsible artificial‑intelligence policy that aligns with international standards and human‑rights safeguards. SHARE argues Shopify lags behind peers such as eBay,...
Prakash Arunkundrum, HP’s First-Ever Chief Strategy and Transformation Officer, Bets Edge AI Will Help Companies ‘Bring the Token Cost Down’
HP appointed Prakash Arunkundrum as its first chief strategy and transformation officer to champion edge AI, aiming to slash token‑based cloud costs by running models locally on AI‑PCs and other devices. The company’s AI‑PCs now account for 35% of its...
LinkedIn Expands Access to Crosscheck AI
LinkedIn is rolling out its AI model comparison tool, Crosscheck, to all U.S. users this week. The platform lets professionals submit any prompt and receive side‑by‑side responses from two leading AI models—drawn from OpenAI, Anthropic, Google, Microsoft and others—without revealing...
Joseph Lazzarotti Comments on Workplace Risks Associated with AI Notetakers
Senior communications executive Joseph Lazzarotti warned that AI‑driven notetakers, now common in corporate meetings, could become a company’s biggest liability. Citing a Fortune piece, he highlighted how these tools automatically record, transcribe, and summarize conversations, often without clear employer oversight....
Gartner Survey Shows AI-Enabled Next Best Actions Lift Sales Growth 2.6‑Fold
Gartner’s latest survey of 227 chief sales officers reveals that organizations that give sellers AI‑enabled next‑best‑action guidance are 2.6 times more likely to achieve commercial growth. The study also links AI upskilling to a 2.4‑fold boost in revenue growth and highlights...
Microsoft and EY Link Up to Speed Enterprise AI Adoption
Microsoft and EY announced a joint investment of more than $1 billion over the next five years to accelerate enterprise AI adoption. The partnership will deploy Microsoft engineers and EY consultants to build industry‑specific AI solutions for finance, tax, risk, HR...
Google Rolls Out Gemini 3.5 Flash, Accelerating AI Agents for Enterprise Workflows
Google unveiled Gemini 3.5 Flash, a new AI model designed to run faster and handle coding and agentic tasks more effectively. The launch targets enterprise customers seeking to embed AI agents directly into business processes, but analysts warn that real‑world...
Dassault Systèmes Launches AI‑Powered HomeByMe Kitchen Suite to Transform Retail Design
Dassault Systèmes introduced the HomeByMe Kitchen Expert Suite, an AI‑enabled, cloud‑based platform that unifies 3D kitchen design, sales management and manufacturer catalogues. The solution aims to streamline omnichannel workflows and lift web‑to‑store quote conversion rates for kitchen retailers.
Juicebox Rolls Out Autonomous AI Recruiting Agents, Claiming Fivefold Efficiency Boost
San Francisco‑based Juicebox App unveiled Juicebox Agents, autonomous AI recruiters that continuously source candidates from over 30 public data sources. Early adopters report up to a fivefold increase in recruiter efficiency and a 50% reduction in sourcing time, addressing the...
Cornerstone Unveils AI‑Driven Platform Reinvention to Transform Corporate Learning
Cornerstone, the $1 billion‑revenue leader in corporate learning technology, announced a comprehensive AI‑powered redesign of its platform. The move targets its 7,000 enterprise customers and 140 million users as the industry shifts away from traditional course‑centric models.
Tripster Launches in Scotland, Promising AI‑Driven Overhaul of Hotel and Short‑Stay Management
Tripster, a Scottish startup, has launched a cloud‑based platform that uses AI and analytics to streamline bookings, pricing and operations for hotels and short‑stay operators. The service aims to lift profitability and guest experience as Scotland’s tourism sector grapples with...
Google Rolls Out AI Ad Agents, Sparking Marketer Trust Concerns
At Google Marketing Live, the company introduced a suite of Gemini‑driven AI ad agents, including the “Ask Advisor” tool that can set up campaigns and surface insights automatically. Marketers responded with caution, citing opaque decision‑making and a need for manual...
Pentagon Picks Shield AI to Add Swarm Pilot to Low‑Cost LUCAS Drone
The Pentagon has selected Shield AI to integrate its Hivemind swarm software into the Low‑Cost Uncrewed Combat Attack System (LUCAS) drone. The AI pilot will let dozens of inexpensive drones coordinate in real time, a capability the service plans to...
Prelude Secures $20 Million Series A to Build AI‑Driven Trust Platform
Paris‑based Prelude closed a €17.2 million ($20 million) Series A round led by 20VC, adding investors such as Singular, Seedcamp, Deel and FDJ Ventures. The funding will fuel the rollout of its new Auth and Intel APIs, aiming to cut verification costs...
WisdomAI Launches Autonomous Analytics Agents to Automate Enterprise Workflows
WisdomAI unveiled its Analytics Agents, AI‑driven tools that not only generate insights but also execute approved actions across enterprise data stacks. The agents, built on the company’s Federated Agentic Intelligence platform, aim to automate routine business processes while preserving auditability...
AI‑Designed Protein Nanocages Reach 220 Nm, Paving Way for Next‑Gen Vaccines
Researchers from Pohang University of Science and Technology and the University of Washington announced AI‑guided creation of quasisymmetric protein nanocages ranging from 70 nm to 220 nm. Published in Nature, the work shows a single, computationally designed protein can self‑assemble into virus‑like...
StoreClaw Launches AI Growth Engine to Automate E‑commerce Acquisition, Conversion and Retention
StoreClaw introduced its AI Growth Engine, a cross‑platform SaaS solution that moves beyond suggestion‑only tools to execute acquisition, conversion and retention actions for e‑commerce sellers. The platform combines AI reasoning with pre‑built “Skills” that act as playbooks, giving merchants a...
SalesboxAI Unveils AI‑Native GTM Platform to Replace Fragmented Revenue Stacks
SalesboxAI announced a unified, agentic, signal‑driven go‑to‑market platform that consolidates 15‑25 disparate tools into a single AI‑native system. The launch targets B2B revenue teams seeking faster pipeline generation and higher conversion rates.
Who Owns a Song No One Wrote?
Generative‑AI music tools can produce chart‑topping songs, but U.S. copyright law still requires a human author. Recent cases such as Thaler v. Perlmutter reaffirm that AI outputs themselves cannot be owned, and prompts rarely meet the originality threshold for protection....
From AI Ambition to Health System Execution: Closing the Gap Between Promise and Practice
Health systems are pouring capital into artificial intelligence, yet up to 80% of pilots never leave the testing phase and roughly 95% fail to generate a return on investment. To address this gap, the Digital Medicine Society and Qualified Health...
AI Message Writer
AI message writers—powered by large language models—are being embedded directly into platforms like Slack to draft, edit, and summarize workplace communications. By leveraging real‑time conversation context, these tools can adjust tone, length, and format, delivering drafts that feel native to...
Relational Foundation Models for Enterprise Data with Jure Leskovec - #768
In this episode, Sam Charrington talks with Jure Leskovec, co‑founder and chief scientist of Kumo and a Stanford professor, about his work on relational foundation models that can reason directly over structured enterprise data without any task‑specific training. Leskovec explains...
SVG Rewind: ‘AIQ: Where Data Meets Dirt’ Technology Brings Real-Time Data Analysis to Rodeo
The Cowboy Channel and Teton Ridge unveiled AIQ, a video‑based AI system that captures real‑time data on rodeo events such as bull riding, saddle bronc, bareback, and barrel racing. Developed with TWG AI, Palantir and NVIDIA, the platform creates skeletal...
Aetina Demonstrates Edge AI Platforms for Robotics, Vision AI and Enterprise Automation at COMPUTEX 2026
Aetina will exhibit its next‑generation edge AI portfolio at COMPUTEX 2026, featuring NVIDIA‑Jetson‑based robotic arms, vision AI modules, and on‑premise large language model workstations. The showcase highlights integrated Physical AI that combines voice, vision, reasoning and motion control without cloud...
Context Layer Could Curb Rising AI Inference Costs
AI inference costs are going to be a big concern: What's the fix? https://t.co/s6mQvdYK7b AI inference costs are turning up on earnings calls as enterprises try to get a handle on token costs as they ramp AI agents. The bigger...
AI Model Autonomously Refutes 80-Year-Old Math Theory
A reasoning model just autonomously disproved an 80-year-old famous math theory, in what the company is calling a first for AI in the field https://t.co/EzGXCMGoqG #ArtificialIntelligence #Innovation #Technology #Tech #TechNews https://t.co/mLU2CYHIEC
A 0.12% Parameter Add-On Gives AI Agents the Working Memory RAG Can't
Researchers introduced delta‑mem, a lightweight memory module that adds only 0.12% of a language model’s parameters yet outperforms much larger alternatives on long‑context benchmarks. By compressing interaction history into a fixed‑size matrix and updating it with a gated delta‑rule, the...
Agents Get Phone Numbers for AI-Driven Real-Time Calls
We just added the ability for your agents sign-up for phone numbers and make AI-native phone calls using Circle Agent Stack and USDC. Get your agent a phone number on Twillio and make AI voice-native real-time calls with BlandAI....
Even Top AI Hallucinates Hardware Details, Lacks Self-Awareness
Vibecoding hallucinations with a very confused anthropic/claude/opus trying to describe what is going on on my MacStudio. Inside an AI that can write code, misread top, and hallucinate about your hardware all at once. State‑of‑the‑Art, but zero self‑awareness: if this...
T-Mobile's Network-Native AI Live Translation Beta Is Here, and This Is Just the Start
T‑Mobile has launched a beta of its network‑native AI Live Translation, accessible by dialing *87* during a call. The service runs directly on T‑Mobile’s 5G edge and distributed cloud, requiring no app download and working on virtually any handset, from...
Taiwan Is Cracking Down on AI Chip Smuggling
Taiwanese prosecutors are seeking detention of three individuals accused of forging paperwork to ship about 50 Super Micro servers, equipped with Nvidia GPUs, to China. The hardware falls under U.S. export restrictions introduced in 2022 to curb China’s access to...
Trump Postpones Executive Order Focused on AI Security
President Donald Trump announced a postponement of an executive order that would have created a 90‑day voluntary testing and vetting regime for frontier AI models. The draft order would have empowered the NSA, Treasury and several cybersecurity agencies to evaluate...
GPT‑5.2 Matches Top Human Peer Reviewers in Study
Seems GPT-5.2 reaches expert level in peer review: 45 scientists took 469 hours evaluating human & AI reviews on 82 papers. "Surprisingly, current AI reviewers are competitive even with the top-rated reviewers in Nature’s official peer review..." though not without weaknesses....
JPMorgan Begins Global AI Rollout After Dimon Signals Hiring Shift
JPMorgan is deploying artificial‑intelligence tools across its global investment‑banking division, becoming one of the first banks to roll out the technology at scale. CEO Jamie Dimon announced a hiring shift toward AI specialists while cutting back on traditional bankers. The...
Can a Sparse-AI Hardware Architecture for Data Centers Work?
Weight‑pruning and activation sparsity can compress neural networks up to 50× and reduce compute by 2‑10×, creating a theoretical two‑order‑of‑magnitude MAC reduction. The Sparse Computing Core Technology (SCCT) architecture leverages a multibus design with private SRAM per lane, offering three...
Why the World’s Banks Are so Worried About Anthropic’s Latest AI Model
Anthropic’s new AI model Mythos has uncovered thousands of zero‑day vulnerabilities across major operating systems and browsers, prompting banks worldwide to sound the alarm. The company has limited access to a defensive coalition of about a dozen partners—including Microsoft, AWS,...
Predicting Cancer Outcomes with a Selfie
Harvard researchers have unveiled a second study confirming that an AI‑driven facial analysis tool, FaceAge, can predict cancer survival outcomes. The work shows patients who appear younger than their chronological age, or whose facial aging slows during treatment, enjoy significantly...
AI and the Future of Work: From Preliminary Findings to National Action
The Special Competitive Studies Project’s Task Force on AI and the Future of Work released a Preliminary Findings report at the AI+ Expo, outlining how artificial intelligence is rapidly reshaping the U.S. labor market. The report notes AI’s diffusion is...
The Times Introduces AI-Generated Audiences for Better Ads
The Times has launched a product that allows advertisers to use AI-generated audiences to improve ad effectiveness. They laid out how it works to me for The Current https://t.co/UXFE7hq9qk
Nvidia Earnings Show AI Spending Moving Beyond GPUs
Nvidia reported record fiscal 2027 Q1 revenue of $81.6 billion, driven by a 92% surge in data‑center sales to $75.2 billion. The company announced a new reporting structure that separates its business into Data Center and Edge Computing, with the Data Center split...
How AI Will Change (Healthcare) Jobs | Out-Of-Pocket
Basata secured a $21 million Series A to deploy AI agents that answer calls, schedule appointments, and process referrals for specialty practices, serving over 500,000 patients nationwide. The article argues that healthcare is ahead of other sectors in adopting AI‑driven work surveillance,...
James Harden, Carmelo Anthony & the AI-Powered Hollywood Pivot
James Harden released an AI‑generated, anime‑style short film on Instagram, reaching his 11.9 million followers and sparking intense fan debate. The clip was produced in under a week using AI‑native workflows, demonstrating how athletes can create high‑impact entertainment without traditional studio...
Trump Delays AI Security Executive Order: ‘I Don’t Want to Get in the Way of that Leading’
President Donald Trump postponed signing an executive order that would mandate pre‑release security reviews of advanced AI models. He cited dissatisfaction with the order’s language, especially provisions requiring companies to share models with the government 14 to 90 days before...
Text Aims to Turn Customer Service Into A Profit Engine with New Agentic AI Capabilities
Text unveiled AI selling agents and custom workflow tools that turn customer service chats into revenue‑generating interactions, especially for Shopify merchants. The agents can detect intent, recommend products, qualify leads and close sales, while an AI Supervisor program helps teams...
The Agentic AI Platform Wars
The article warns that AI is reshaping enterprise software markets, prompting vendors to vie for the role of a unified, agentic AI orchestration layer. Companies across verticals, especially CX and contact‑center organizations, face pressure to choose a platform before the...
Meta Is Spending Billions on AI While Cutting Thousands of Jobs — And Silicon Valley Workers Are Panicking
Meta announced a roughly 10% workforce reduction, cutting about 11,000 jobs, while simultaneously committing billions of dollars to AI chips, data centers, and large‑scale computing infrastructure. The layoffs are part of a broader 2026 tech downturn that has already eliminated...