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:
Hong Kong Productivity Council Partners With Schools to Expand AI Education
Hong Kong Productivity Council (HKPC) has forged partnerships with eight academic organisations to roll out AI education across more than 500 primary and secondary schools. The "AI Empowerment: Future Skills" programme, built on HKPC’s "AI for All" training, delivered over 1,100 activities to 60,000 participants in 2025‑2026, with 350 focused on AI. New memoranda of understanding will provide AI health checks, ethics‑focused curricula, teacher training and industry immersion. A pilot at Lee Kwok Po Secondary School already offers AI literacy for teachers, students and parents, while a 2026 Innovation and Technology Study Tour will showcase the initiative further.
Anthropic Claude Opus 4.8: Technical Architecture, Capabilities and Implications for Healthcare Technology
Anthropic unveiled Claude Opus 4.8 on May 28 2026, a frontier AI model built on a $65 billion Series H round that lifts its post‑money valuation to $965 billion and drives annualized revenue past $47 billion. The model delivers higher tool‑mediated reasoning (57.9%) and computer‑use scores (83.4%),...
One Company Reportedly Spent $500 Million on Claude in One Month After Failing to Cap AI Usage
An unnamed firm allegedly burned through $500 million in a single month after failing to cap usage of Anthropic’s Claude model. The incident underscores a broader industry trend where flat‑rate AI contracts mask soaring per‑request costs, prompting executives like Microsoft’s and...
NIST Expands AI Consortium’s Scope, Calls for New Members
The National Institute of Standards and Technology has renamed its AI Safety Institute Consortium to the NIST Artificial Intelligence Consortium and broadened its mission to cover AI measurement, innovation and adoption. The revamped group will focus on building an AI...
9 Demos of Gemini Omni and Gemini 3.5 in Action
At Google I/O 2026 the company introduced Gemini Omni, a multimodal model that can generate and edit video using text, image, audio or video inputs, and Gemini 3.5 Flash, an agentic model built on the Antigravity harness that excels at...
After Nvidia’s $20B Not-Acqui-Hire, AI Chip Startup Groq Reportedly Raising $650M
AI chip startup Groq is pursuing a $650 million funding round from its existing investors to accelerate its inference‑cloud services. The round follows a December “not‑acquisition” deal with Nvidia that paid Groq shareholders $20 billion in cash and transferred senior staff and...
Malaysia Highlights AI Role in Integrated Border Security, Cybercrime Legislation
Malaysia is embedding artificial intelligence at the core of its upcoming border security framework, linking radar, satellite and on‑site sensors across land, sea and air. The National Border Security Control Master Plan 2026‑2030 will launch an AI‑driven surveillance system that offers...
Grady Health System Goes Live with Sectra Amplifier Services to Enable Smooth Implementation of AI—Enhancing Diagnostic Workflows Across Sites
Grady Health System, one of the largest U.S. safety‑net providers, has gone live with Sectra Amplifier Services, a cloud‑based AI‑as‑a‑Service platform that streamlines deployment of diagnostic algorithms. The rollout includes Avicenna.ai’s CINA Chest and Therapixel’s Mammoscreen, enabling faster, more accurate...
Mustafa Suleyman's Case Against Open-Source AI Shortcuts
Microsoft AI chief Mustafa Suleyman warned that relying on distillation—training smaller models on data generated by larger frontier labs—creates a dead‑end for enterprises. He says Microsoft is building its next‑generation models with "zero distillation" to avoid the hidden biases and...
Zig President Says AI Coding Contributions Are 'Invariably Garbage,' So He Banned Them
Zig, the open‑source programming language, has instituted a strict ban on any code, edits, or debugging that originates from large language model (LLM) tools. President Andrew Kelley described AI‑generated pull requests as "invariably garbage" because they add no value and...
Sweden Authorizes Police Use of Live Facial Recognition
Sweden's parliament approved a law allowing police to deploy live facial recognition (LFR) in cases of kidnapping, human trafficking, serious crimes and imminent threats to life. The measure, slated to take effect on July 1, 2026, requires court authorization except...
AI Slop Is Coming for Your Playlists
AI‑generated songs that mimic the 2019 reggae track “Angels Above Me” have gone viral, racking up millions of streams on Spotify, TikTok and iTunes charts in Germany and Austria. Platforms like Spotify report removing over 75 million spammy tracks and are...
Salesforce Dreamforce San Francisco 2026
Salesforce’s Dreamforce 2026 will convene from September 15‑17 at San Francisco’s Moscone Center, gathering thousands of AI, data and CRM professionals. The three‑day conference will feature more than 1,600 expert‑led breakout sessions, keynotes from CEO Marc Benioff, and a deep...
EPlus Sees FY26 Sales Boom As AI Is ‘Becoming Embedded’
ePlus reported double‑digit growth in FY26, with gross billings reaching $3.8 billion, a 17% increase, and net sales climbing 22% to $2.4 billion. Adjusted EBITDA surged 49.5% to $204.8 million, reflecting strong demand for AI‑driven infrastructure, security and cloud services. CEO Mark Marron...
We Asked the ‘Future of Truth’ Author to Explain How He Used AI. It Didn’t Go Well
Steve Rosenbaum’s new book *The Future of Truth* examines how artificial intelligence reshapes perception, but a recent Wired excerpt revealed fabricated or misattributed quotes. The New York Times identified more than half a dozen such errors, prompting Rosenbaum to acknowledge that a...
Why Regulatory Rigour Is Becoming a Competitive Edge in Clinical AI
Heidi Health, a Melbourne‑based health‑tech firm, is turning regulatory rigor into a market advantage as it expands across Canada. By embedding legal, compliance and former clinicians into its core product team, the startup secured placement on Ontario’s AI Scribe Vendor...
Cognition’s Scott Wu Says AI Coding Agents Shouldn’t Replace Humans
Cognition, the creator of the AI coding agent Devin, closed a $1 billion Series D round, valuing the startup at $26 billion. The company reports that Devin authored 89% of its code, positioning the tool as a self‑driving software development partner rather than...
Samsung Ships First 12‑Layer HBM4E Samples, Boosting AI Memory Bandwidth
Samsung Electronics started shipping the industry's first 12‑layer HBM4E memory samples, offering 48 GB capacity and up to 3.6 TB/s bandwidth per stack. The launch positions Samsung ahead in the high‑bandwidth memory market as AI and high‑performance computing demand surge.
DNS-AID Will Make AI Agents Easier to Discover, Says Linux Foundation
The Linux Foundation announced DNS‑AID, a proposal to repurpose the existing Domain Name System as a global, vendor‑neutral directory for AI agents. By defining a well‑known subdomain format (_index._agents.{domain}), the standard lets agents discover and verify each other without new...
Dell Posts Record $43.8B Q1 Revenue, Up 88% on Enterprise Demand
Dell Technologies announced a record $43.8 billion first‑quarter revenue, an 88% increase year‑over‑year, powered by surging AI server and traditional enterprise hardware sales. The company lifted its fiscal‑2027 revenue guidance by roughly $27 billion and warned that component shortages could curb growth...
Micron's $1 Trillion Valuation Signals Structural AI Memory Demand
Micron Technology’s market cap topped $1 trillion as its high‑bandwidth memory chips fuel a structural surge in AI hardware demand. The rally, backed by hyperscale AI spend and investor backing from Anthropic, confirms that AI‑driven memory needs will shape consumer devices...
AI-Fueled Earnings Lift S&P 500 10% YTD, but Consumer Sentiment Raises 2026 Questions
First‑quarter results from Alphabet and Nvidia helped push the S&P 500 up almost 10% this year, underscoring AI’s outsized impact on large‑cap earnings. Yet fading consumer confidence and rising inflation introduce uncertainty for the market’s 2026 trajectory.
Channel Brief: MSPs Are Moving Past the AI Add-On Conversation
The managed‑service provider (MSP) market is moving AI from a bolt‑on feature to a native layer embedded in platforms, security tools, and data systems. MSP platform vendors are weaving AI into daily workflows, making AI a foundational element of service...
Gemini Architects Reveal Vision Behind AI Frontier
Hear the architects of Gemini reflect on their journey to continue pushing the frontier of AI, on this episode of Release Notes. @JeffDean, @koraykv, @OriolVinyalsML, and @NoamShazeer sit down on camera together to share a behind-the-scenes look at the people behind...
Waymo Expands Fleet and Mapping Footprint as Autonomous Ride‑Hailing Gains Traction
Waymo announced the rollout of its new Ojai robotaxi in three California markets and began extensive mapping work in Virginia, while its existing service now logs about 500,000 rides a week across 11 U.S. metros. The moves highlight the company’s...
How Drones And AI Are Mapping Heat Loss Across Neighbourhoods in the UK
Kestrix, founded by Forbes 30 Under 30 honoree Lucy Lyons, is using drones equipped with thermal imaging and artificial intelligence to produce detailed 3‑D heat‑loss maps of UK neighbourhoods. The system, described as the “Google Maps of heat loss,” lets...
AI Growth Exposes Gaps in Governance and Readiness
Veeam’s new study of 300 technology and business leaders across finance, healthcare, government, manufacturing and tech finds that AI deployment is now near‑ubiquitous, with 95% of organizations using the technology in some form. Yet the rapid rollout outpaces governance, with...
Pennsylvania Invests $30M in AI Traffic Signals to Unclog Sports Complex
Pennsylvania Governor Josh Shapiro announced a $30 million investment that will install AI‑controlled traffic signals and a new I‑76 ramp to alleviate gridlock around Philadelphia’s Sports Complex. The project, part of a broader $30 million infrastructure push, aims to serve the 378...
Why AI Starts With Better Data, with Parijat Banerjee.
In this episode of the Digital Banking Podcast, Parijat Banerjee, Global Business Head for Financial Services at Latent View Analytics, explains why AI’s real power starts with high‑quality data. He traces the evolution of data storage—from costly floppy disks to...
Snyk Creates Operational Roadmap for the AI Governance Maturity Model
Cybersecurity firm Snyk released an executive guide that outlines a five‑phase operational roadmap for AI governance, embodied in its Evo platform. The model moves organizations from static, paper‑based policies to continuous discovery, risk scoring, real‑time enforcement, least‑privilege controls, and ongoing...
New AI Search Optimization Checklist Streamlines Validation and Reporting
🚨 I’ve just updated my AI Search Optimization Checklist (and worksheet) 👇 Following a workflow for a strategical AI search optimization process: ✅ Which prompts and journeys do we actually want to influence? ✅ Where are we visible, cited, recommended or missing? ✅ Which...
AI-Driven CSM Pairs Humans with Autonomous Partners
Digital CSM: Revolutionizing Customer Success with AI https://t.co/YU7ykl1oS8 The future of Customer Success isn’t “more humans doing more tasks.” It’s humans + autonomous partners delivering outcomes together. @castdotapp https://t.co/nYkZIfaKTy
Your AI Isn’t My AI: The Quiet Splintering Ahead
Large language models are moving from a single, dominant platform toward a fragmented landscape shaped by national interests, open‑weight ecosystems, and autonomous agents. Countries such as China, India, France and a trans‑Atlantic coalition are building sovereign LLMs that embed local...
Tesla FSD Completes 3,760‑mile Autonomous Canada Drive
JUST IN: Tesla FSD completed a 3,760 mile drive across Canada with zero human input, the longest autonomous drive ever recorded
Design Your Dream Home in Seconds with AI
It’s never been easier to design your dream house. Draw a shape. Define your rooms. Set your constraints. @DraftedAI generates complete floor plans, elevations, and 3D home designs in seconds. Over the last month, 120,000 people generated 325,000+ home designs with https://t.co/XqC0LP5n3y.
AI Prices Are Going Up, Up, Up – And What This Means For Enterprise AI
The article warns that enterprise AI tool costs are set to surge as the underlying technology remains costly to develop and operate. Rising fees will pressure organizations to prioritize high‑value use cases and could accelerate consolidation among AI vendors. While...
OpenAI Retires Last GPT-4, Ending AI Era
OpenAI just quietly retired the final GPT-4 model from ChatGPT — closing the chapter that changed AI forever https://t.co/K5YiivRA53
AI Prices Are Going Up, Up, Up – And What This Means For Enterprise AI
AI infrastructure spending is accelerating toward a $1 trillion annual run‑rate by 2026, with the Big 4 hyperscalers alone committing $500‑$750 billion to data‑center capacity. Total AI‑related capital outlays could approach $1 trillion in 2026 and $6.3 trillion by 2030, according to Gartner. As investors...
ByteDance Builds AI Chips Rivaling Nvidia’s Groq
China’s ByteDance Developing New AI Chips Like Those from Nvidia Partner Groq — The Information https://t.co/PtVHdCSL3B
Fanuc, Google Advance Industrial Robotics as Part of Recent AI Deals
Industrial robot maker Fanuc teamed with Google to embed advanced AI across its robot portfolio, from 3‑kg payload units to multi‑ton machines, leveraging Google’s ROS contributions via Intrinsic. Kawasaki Heavy Industries opened a Silicon Valley physical‑AI center, collaborating with Nvidia,...
Gartner Says CFOs Gain a Competitive Advantages From Strategic AI Deployment, Not AI Spending Levels
Gartner’s latest research shows that CFOs who embed AI strategically across product, sales, marketing and customer growth achieve superior revenue, margin and return‑on‑capital results, while merely increasing AI spend does not guarantee better outcomes. In a study of 101 efficient‑growth...
Anthropic Opus 4.8 Shows the AI Lab Is Paying Attention to Customers
Anthropic closed a $65 billion Series H round, lifting its valuation to $965 billion, and simultaneously launched Claude Opus 4.8. The new model adds dynamic workflows, a fast‑mode that is three times cheaper, and enhanced honesty flags for hallucinations. Opus 4.8 is positioned as a...
How Has Anthropic Become the Top AI Company?
Anthropic’s revenue exploded after launching Claude 4.5, which prompted customers to upgrade and double usage. Token consumption tripled while costs rose modestly, delivering a twelve‑fold revenue increase and pushing ARR to $47 billion, with a target of $100 billion. The company recently closed...
Weekend Essay: Why ChatGPT Is Unsettling the Advice Profession
OpenAI has launched a ChatGPT personal‑finance tool for U.S. users that lets Pro subscribers link accounts at more than 12,000 banks and brokerages via Plaid. The announcement sparked a split in the UK advice community, with some seeing AI as...
Why Generic LLMs Fall Short for Critical Engineering Documentation
Engineering documentation demands absolute correctness, traceability, and version governance, which generic large language models (LLMs) cannot guarantee on their own. A new white paper from llmda.ai shows that building the necessary ingestion, grounding, audit, and publishing layers around a model...
NeuroPace Announces FDA Approval of ECoG Assistant™, Advancing AI-Driven Epilepsy Care
NeuroPace received FDA clearance for ECoG Assistant™, its first AI‑driven clinician‑enabled feature for the RNS System. The tool leverages a proprietary dataset of 124,450 epileptologist‑labeled intracranial EEG records to flag ECoGs of interest and present trends in a single view....
Stratview Research White Paper Analyzes AI Shift in Composites Manufacturing
Stratview Research released a free white paper examining AI’s role in composites manufacturing. The study shows AI can cut automated fiber placement waste by 80%, speed mold design by 80%, and achieve 96% first‑time‑right yields. A survey of over 25...
Microsoft Slaps New Coat of Paint on Copilot, Buries Annoying Button
Microsoft refreshed its Copilot experience in Microsoft 365, adding a new visual design, a task‑aware prompt workspace, and an option to move the floating Copilot button back to the ribbon. The update claims the app loads more than twice as...
Maybe We Should Pretrain on Synthetic Data About Good-but-Reward-Hacking AIs
The post proposes "inoculation pretraining," a hybrid of inoculation prompting and alignment pretraining that injects synthetic data about "good‑but‑reward‑hacking" AI personas into the pretraining corpus. By boosting the prior probability of benign reward‑hacking behaviors, the approach aims to prevent emergent...
Claude Opus 4.8 Is Now Available in Microsoft Foundry
Anthropic has released Claude Opus 4.8 on Microsoft Foundry, giving developers and enterprises access to its most capable model for coding, agentic tasks, and deep enterprise reasoning. The upgrade focuses on longer context handling, more reliable tool use, and consistent...