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:
Workday Rolls Out Sana, Its Conversational AI Gateway to Enterprise Work
Workday announced the general availability of Sana, an AI‑powered conversational interface that replaces traditional menus for its HR and finance applications. The launch includes three components: Sana for Workday, the Self‑Service Agent that automates routine queries and complex workflows, and Sana Enterprise, which connects to dozens of third‑party apps. The solution runs within Workday’s security and identity framework and is offered via usage‑based Flex Credits, shifting pricing toward outcome‑driven metrics.
AI-Generated Text Dilutes Writing's Signal of Thought
LLMs can write. The prose is often smooth and the structure is darn convincing. And when language can be generated that easily and at that scale, writing begins to lose some of what made it feel substantial in the first...
AI Misses Nuance: Human Insight Still Essential
I'll never forget the moment a founder told me, "I'll just take these transcripts and tell ChatGPT to spit out jobs to be done." We were in the middle of a jobs-to-be-done research project. We'd mapped four or five customer...
Telcos Are the Best Channel to Democratize AI
The blog argues that while only a fraction of humanity currently interacts with frontier AI, the real bottleneck is distribution, not model size. It posits that the economic unit of AI has shifted to the inference token, which will be...
FANUC Accelerates Physical AI in Industrial Robotics, Leveraging NVIDIA Technologies
FANUC announced a strategic partnership with NVIDIA to accelerate "physical AI" in its industrial robot lineup. The collaboration embeds NVIDIA Jetson edge modules, Isaac Sim, and Omniverse into FANUC's robots and ROBOGUIDE software, enabling photorealistic digital twins and real‑time AI...
Introducing Sana From Workday: Superintelligence for Work That Finds Answers, Takes Action, and Automates Workflows
Workday unveiled Sana, a superintelligence platform that embeds generative AI into its cloud suite. Sana can surface answers from enterprise data, execute actions on behalf of users, and automate routine workflows across HR, finance, and planning functions. The solution leverages...
Inclusive AI Over Speed: Sweden's Call at Delhi Summit
Sweden’s deputy prime minister Ebba Busch called for a more inclusive approach to AI at the India AI Impact Summit in New Delhi. She argued the next phase of AI development should focus on accessibility and solving real-world problems. The bigger question...
Orchid Security Recognized by Gartner® as a Representative Vendor of Guardian Agents
Orchid Security has been named a Representative Vendor in Gartner’s inaugural Market Guide for Guardian Agents, which focuses on managing identities and access for AI agents with zero‑trust policies. The guide warns that AI agents expand “identity dark matter,” creating...
AI Agents in Health Care: What They Say when We Aren’t Listening
Moltbook, a Reddit‑style platform for autonomous AI agents, has become a live laboratory where "moltbots" discuss health, medicine, and human well‑being without human moderation. By February 2026, over 1,000 posts referenced human health, revealing three dominant themes: AI envisioning its...
Gamma Adds AI Image-Generation Tools in Bid to Take on Canva and Adobe
Gamma has unveiled Gamma Imagine, an AI‑driven image‑generation tool that lets users create brand‑specific marketing assets such as charts, infographics, and social graphics from simple text prompts. The product plugs into a suite of integrations—including ChatGPT, Claude, Zapier, Atlassian, n8n,...
Nvidia's Agentic AI Stack Is the First Major Platform to Ship with Security at Launch, but Governance Gaps Remain
Nvidia unveiled its agentic AI stack at GTC, marking the first major AI platform to ship with security baked in rather than added later. Five security vendors—CrowdStrike, Palo Alto Networks, JFrog, Cisco, and World Wide Technology—each cover a distinct layer...
A Modern Clerk’s Office: Lessons From Dallas County
In this episode, Dallas County Clerk John Warren discusses the county’s journey from paper‑based record keeping to a fully digital, AI‑enhanced clerk’s office. He explains how early resistance was overcome by demonstrating efficiency gains, such as simultaneous access to documents,...
The AI Bubble, Like the Housing Bubble, Is a Big Problem and It’s Not Complicated
Dean Baker warns that the current surge in AI company valuations mirrors the speculative excesses of the early‑2000s housing boom. He argues that inflated expectations are driving a classic bubble, with venture capital pouring billions into startups that lack sustainable...
AI Empowers Elite Workers to Replace Mediocrity
The cost of being average has never been higher. It's no longer a question. AI helps businesses eliminate mediocre employees. But it's not that AI is replacing workers. It's that ultra-talented team members (now armed with significantly more leverage) can play the role 5...
Nvidia Unveils DGX Spark with Grace Superchip
Signed $NVDA DGX Spark which features Grace Blackwell superchip (CPU + GPU) & SK Hynix LPDDR5X DRAM memory
Chip Wafer Shortage Will Run Through 2030 as AI Demand Overwhelms Supply: SK Hynix Chief
SK Hynix chairman Chey Tae‑won warned that a global wafer deficit exceeding 20% will persist until at least 2030, driven by soaring AI‑related high‑bandwidth memory (HBM) demand. He said capacity expansion will take four to five years, and SK Hynix,...
Tongyi Lab Releases First Open-Source Multi-Speaker AI Dubbing
"AI dubbing? That's easy, right? Just match the voice to the lips." That's what most people think. But here's the thing: multi-speaker dubbing is actually ONE OF THE HARDEST problems in AI video. And Tongyi Lab just solved it. World's FIRST OPEN-SOURCE AI dubbing...
The Download: OpenAI’s US Military Deal, and Grok’s CSAM Lawsuit
OpenAI has struck a controversial agreement granting the Pentagon direct access to its generative AI models, with officials hinting the technology could soon be used to assist target selection in the Iran theater and integrate with Anduril’s drone systems. Simultaneously,...
Trustpilot Partners with AI Companies as Traditional Search Declines
Trustpilot is forging partnerships with major eCommerce platforms to feed AI shopping agents with its review data. CEO Adrian Blair said AI‑driven search generated a 1,490% surge in click‑throughs and positions the company to hit a 30% operating margin by...
Chat and Voice Will Remain Core AI Interfaces
Every month I periodically see the recycled take that “something better than chat” is coming for AI. That chat is temporary. In fact, I predict the opposite. More of our work and life will happen through chat and voice interfaces of...
Data Meets Dyslexia: What AI Could Mean for Identification in Schools
Artificial intelligence is poised to become a cornerstone of modern classrooms, and educators are exploring its role in dyslexia identification. AI‑driven analytics can sift through reading and performance data to spot patterns that traditional assessments may miss, offering a more...
The Evolution From Prompt Engineering to Concept Engineering
Prompt engineering once unlocked rapid value from large language models, but its reliance on fragile, monolithic cues creates brittleness, hidden requirements, and token bloat. Concept engineering reframes interactions as explicit contracts, modular components, and measurable metrics, turning prompts into...
GitGuardian Reports an 81% Surge of AI-Service Leaks as 29M Secrets Hit Public GitHub
GitGuardian’s 2026 State of Secrets Sprawl report reveals an 81% year‑over‑year surge in AI‑service credential leaks, pushing total exposed secrets on public GitHub to roughly 29 million. AI‑assisted coding, exemplified by Claude Code, shows a 3.2% leak rate—about twice the platform...
Future Is Training Its AI On Publisher First-Party Data
Future Media unveiled Helix, an AI‑powered audience platform that builds on its first‑party data assets. Helix replaces the legacy Aperture system in the U.S., adding predictive modeling and real‑time audience segmentation for outcome‑based campaigns. Early tests across 20 fashion, retail...
Phishing Emails Target AI Defenses with Unique Obfuscation
Cybercriminals are deploying a new email obfuscation method to bypass NLP‑based phishing filters. The tactic inserts hundreds of line breaks and large blocks of benign graymail or random text after the malicious payload, diluting the malicious signal and inflating email...
How AI Tutors Improved Learning in Nigeria
A randomized controlled trial in nine public secondary schools in Edo State, Nigeria tested a six‑week after‑school program where students used GPT‑4 under teacher supervision. Participants achieved learning gains of about 0.3 standard deviations in English, equivalent to roughly 1.5...
AI Policy in Developing Countries
Umar Saif argues that data availability and political economy, not technology, are the primary bottlenecks for AI adoption in developing countries. Pakistan exemplifies a "data desert" where bureaucratic silos impede useful AI applications such as crop‑yield forecasting, education, and health...
‘We Finally Have the Tools to Harness and Scale Law Firm Knowledge’
A Global Legal Post webinar highlighted how generative AI is unlocking firm‑wide knowledge for legal drafting, allowing even small practices to create sophisticated templates without large knowledge‑management teams. Panelists from Gleiss Lutz, Hengeler Mueller and Pérez‑Llorca explained that AI now drafts complex...
An AI Image Generator for Non-English Speakers
Researchers at the University of Amsterdam have unveiled NeoBabel, an open‑source text‑to‑image model that generates pictures directly from prompts in six languages—English, French, Dutch, Chinese, Hindi and Persian. By training on a newly expanded 124 million image‑label dataset and using multilingual...
China AI Boom Peaks; Baidu Leads Profitably
The AI craze in China has peaked - Minimax briefly surpassed Baidu in market cap and is about the same now. Zhipu - very close. What is interesting is that Baidu is actually profitable and has $18B + in revenue....
Advanced Generative AI Course for Engineers – Interview Kickstart Launches New Program Focused on LLM Applications, Prompt Engineering, and Real-World...
Interview Kickstart has unveiled an Advanced Generative AI Course aimed at engineers, data scientists, and AI practitioners. The program dives deep into large language models, prompt engineering, retrieval‑augmented generation, and model evaluation, emphasizing hands‑on project work over theory. Instruction is...
AI‑Sequenced Chatbot Boosts Learning by 6‑9 Months
Interesting paper showing that Bjork’s desirable difficulty works. Tight integration of AI chatbot with ‘intelligent task sequencing’ delivers 'right material at exactly the right level matching the individual’s needs. In other words, AI tutors work better when they do...
AI Speeds Insight, but Can't Replace Human Judgment
A founder fed customer interviews into ChatGPT to extract jobs-to-be-done. Output wasn't wrong. Just goofy. It missed hesitation, energy, contradictions. The tension: AI is useful and easy to misuse. AI-assisted ✅ speeds up your work AI-led ❌ replaces judgment
OpenAI Reportedly Ditches Its "Side Quests" Strategy to Focus on Coding Tools and Business Customers
OpenAI is abandoning its "side quests" approach, consolidating resources around two core pillars: coding tools and business‑focused AI solutions. The shift follows internal criticism that a flood of products—Sora, Atlas, hardware devices, and more—stretched compute and talent thin, leading to...
Roche Launches Biggest Hybrid‑cloud AI Factory with Nvidia
At 3,500+ Blackwell chips, Roche is building the "largest announced hybrid-cloud AI factory" in biopharma with Nvidia: https://t.co/99Eih7EnRD
GPT‑5.4 Handles 5 Trillion Daily Tokens, Hits $1B Revenue
Greg Brockman announced GPT-5.4 is processing 5 trillion tokens per day within one week of launch, exceeding all prior API volume and generating $1B in annualized net-new revenue #ArtificialIntelligence #Innovation #Technology #Tech #TechNews
AI Standardisation Speeds Up
Since SMPTE launched its AI Taskforce in 2020, AI adoption in broadcast has accelerated dramatically, prompting an updated Engineering Report in 2024 that highlights emerging standards. The report recommends standardising on the Model Context Protocol (MCP) and Agent-to-Agent (A2A) frameworks,...
Meta Commits $27B to AI Cloud with Nebius
Meta signed a $27B deal with Nebius to deploy AI cloud infrastructure, including one of the first large-scale deployments of Nvidia's Vera Rubin platform https://t.co/jZIGtPgNBf #ArtificialIntelligence #Innovation #Technology #Tech #TechNews https://t.co/EA3sXbhho4
OpenAI Pivots to Renting AI Servers, Restructures Stargate Team
OpenAI is reportedly restructuring its Stargate computing team, coming alongside a shift in strategy to renting more AI servers rather than building its own data centers https://t.co/GcaUtwhISO #ArtificialIntelligence #Innovation #Technology #Tech #TechNews https://t.co/lt4ueogRAC
Why Adland's Women Are the AI Explorers We Need Right Now
The advertising industry is witnessing a rapid surge in generative AI tools, yet women’s adoption lags sharply behind that of men. Charlotte Bunyan highlights this gender gap, noting that fewer women are experimenting with AI-driven creative workflows. The article argues...
Open‑Source AI Agent Automates Physics Research End‑to‑End
Physical Superintelligence dropped Get Physics Done, an open-source AI agent that can scope research problems, run experiments, verify results, and draft papers https://t.co/0gDAIMmxe7 #ArtificialIntelligence #Innovation #Technology #Tech #TechNews https://t.co/nudT0C4wXm
AI Moves From Emails to Controlling Physical Infrastructure
AI writing emails was just the warm-up. The real disruption came when #AI started running factories, warehouses, and infrastructure. In other words: AI got hands. Great conversation at #SXSW with @shantnu, Nadia V. Gil, and @roopaonline. #Innovation @sxsw https://t.co/tFEVwM0qvh
Who’s Whispering in Your Chatbot’s Ear?
Liberal democracies are increasingly ceding control of AI infrastructure to a handful of private firms, creating a centralized, unaccountable power base. The article argues that algorithmic systems are inherently value‑laden, yet regulators have failed to demand transparency or oversight. This...
AI Marketing Toolkit Grows: New Tools Added
The best list of #AI #marketing tools anywhere continues to expand: https://t.co/1MqMiJP4M3 New tools added recently include @post_everywhere, @HumanticAI, @Jogg_ai, Omnilingual ASR from @Meta, @claudeai Haiku, @hedra_labs, @TrySpiral, @tight_studio, and @averi_ai.
AI's Real Impact Lies in Pricing, Forecasting, Operations
If your AI strategy is drafting emails faster and summarizing meetings, you’re thinking too small. Real value appears when AI touches pricing, forecasting and operations. Anything else is relying on the luck. Happy St. Patrick’s Day. https://t.co/1VaXofchlk #AIStrategy https://t.co/6MsnjSYyT7
Inclusion in AI Answers Is Becoming a Discovery Advantage
Generative AI assistants are reshaping product discovery by delivering synthesized answers that embed brand mentions, creating a new visibility layer beyond traditional search rankings. Similarweb’s 2026 AI Brand Visibility Index shows that a small group of brands dominate AI mentions...
From AI Dependency to Unexpected Disruption
For months, I did 90% of my AI work in ChatGPT. I had dozens of custom GPTs. Hundreds of hours invested. It felt like a finely calibrated machine. Then everything changed. 01/12
TrafficGuard Appoints Head of AI to Battle Rising Fraud
TrafficGuard has appointed Scott Thomson, a former Google, Adobe and Telstra AI leader, as Head of AI to strengthen its fight against sophisticated agentic bots. The move comes as autonomous bots that mimic human behavior are reshaping the digital ad...
What's Missing Between LLMs and AGI - Vishal Misra & Martin Casado
In this episode, Martin Casado interviews Vishal Misra, a Columbia professor who has built a mathematical model of how large language models (LLMs) operate, treating them as massive sparse matrices that update token probabilities via Bayesian-like inference. Misra explains his...
NYU Langone Health’s O’Brien & Major Explain Keys to Effective Nursing/AI Partnership
NYU Langone Health has built a structured partnership between nursing informatics and data science that fuels AI tool development and adoption. Since 2019, the collaboration has produced decision‑support models, a pressure‑injury prevention system, and a fall‑prediction tool, all vetted through...