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:
Reddit's CEO Calls His Company 'the Fuel' For Artificial Intelligence
Reddit CEO Steve Huffman told Mad Money that the platform’s massive archive of human conversations is the "fuel" powering modern artificial‑intelligence models. The company reported a standout quarter, with revenue jumping 69% year‑over‑year to $663 million, daily active users rising 17% to 126.8 million, and gross margins exceeding 90%. Capital spending was a modest $1 million, delivering $311 million in free cash flow, more than double the prior year. Partnerships with Google and OpenAI underscore Reddit’s growing role as a data source for AI developers.
DEAL Desk: Nantum AI Latest Acquisition as OEMs Shift Digital Strategy From Build to Buy
Johnson Controls (JCI) announced it has acquired Nantum AI, formerly Prescriptive Data, to accelerate AI‑driven energy optimization across its smart‑building portfolio. The deal gives JCI an established machine‑learning platform that originated as an in‑house solution for New York real‑estate firm Rudin....
Agentic AI: More Art Than Science, Embrace Uncertainty
code, and the applications they made had clear sharp boundaries. agentic ai is so different. it feels far more like art than science. there is too much complexity, too much fuzziness to really know what works and what doesn’t. you...
Vertical AI Is ONE Of The Most Promising Areas For Investment – Tech Entrepreneur And Investor Serhii Tokarev
Serhii Tokarev argues that vertical artificial intelligence is outpacing horizontal models as the most attractive investment theme in 2026. While foundation models draw 60% of AI funding into a few mega‑rounds, specialized AI solutions are spreading capital across sectors such...
PrairiesCan Invests $7.9 Million in Funding to Support Co.Labs, Local AI Firms
The Government of Canada, via PrairiesCan, pledged $7.9 million CAD (≈$5.9 million USD) to boost Saskatchewan’s tech ecosystem, including $3.7 million CAD (≈$2.7 million USD) for the Co.Labs incubator. The money flows through the Regional Innovation Ecosystems program and will fund Co.Labs’ AgTech and...
Cheap AI Advice Threatens Pricing Landscape for Robo Advisors
We're officially hitting the "robo" stage of AI startups as it pertains to advice, with Astor offering a $39.99/month subscription for access to a 'trained' AI chatbot/voicebot. For advisors, this is reminiscent of the robo-investing phase a decade ago. If you...
Morning Headlines 5/1/26
Larry Ellison is reportedly financing Texas Attorney General Ken Paxton to pursue legal actions against business rivals, a strategy critics say leverages state power for private gain. The lawsuit filed by Paxton’s office has been described by legal analysts as...
Managing AI in Medical Technology: From Innovation to Compliance
AI is rapidly reshaping medical technology, prompting heavy investment in algorithm‑driven devices and software. At the same time, regulators worldwide are tightening oversight, creating a need for structured AI governance. The newly released ISO/IEC 42001 standard offers a lifecycle framework that...
Why ADHD Is the Cheat Code of the AI Era
The post argues that ADHD, long labeled a deficit, is a strategic advantage in the AI era because artificial intelligence now handles linear, repetitive work, leaving pattern‑recognition and rapid idea synthesis as premium skills. Neurodivergent brains excel at juggling multiple...
GPT‑5.5 Matches Mythos Core Capabilities, Lags in Vulnerability Discovery
Is GPT5.5 on par with Mythos for cyber security? Mythos did discover significant vulnerabilities beyond what OpenAI has said about GPT5.5, but that was also the result of substantial harness work outside of Claude, which can be easily replicated. Our...
Govern Your Bots Carefully or Chaos Could Ensue
Gartner warns that Global Fortune 500 enterprises could operate over 150,000 AI agents by 2028, up from fewer than 15 today, creating a risk of uncontrolled "agent sprawl." The firm’s research shows that organizations that adopt strong, centralized governance and...
Why Generic AI Isn’t Enough for Pricing Decisions
The article argues that generic AI tools, while fast and versatile, cannot handle the intricate, high‑stakes pricing decisions of B2B manufacturing and distribution. It highlights that purpose‑built pricing AI integrates real transactional data, governance rules, and system workflows to protect...
Palo Alto Networks To Acquire AI Gateway Startup Portkey
Palo Alto Networks announced it will acquire Portkey, a startup that provides an AI gateway for managing and protecting autonomous agents. The deal, expected to close in the fourth fiscal quarter ending July 31, has undisclosed terms. Portkey’s platform processes trillions...
‘Depth over Breadth’: Health Systems Eye Quality of AI Applications, Not Number
Large U.S. health systems are shifting from counting AI projects to measuring their clinical and operational value. Leaders at Universal Health Services, Advocate Health, UPMC, Mayo Clinic and NYU Langone describe multi‑stage governance that vets pilots for safety, workflow fit...
Solving the Mystery of How Businesses Are Using AI
New research from the NBER shows that firms investing in artificial intelligence grow faster and enjoy higher market valuations, but aggregate productivity gains remain modest. Adoption is uneven, favoring larger companies and those with more high‑skill workers. Early AI tools...
Inside Wealth: Markets Are Underpricing the Risk of Middle East Pullback in AI, Says Tech Investor Jack Selby
Tech investor Jack Selby warns that Middle East sovereign‑wealth funds, which account for roughly a quarter of global AI spending over the next five years, could pull back if the Iran war drags on. A potential retreat could withdraw hundreds...
Your Old-School Process Skills Are a Superpower for Building AI Agents
The article argues that building effective AI agents hinges on two layers—prompting and decision‑logic. While prompt engineering can be picked up quickly, the logic layer requires the kind of conditional workflow thinking that operations, project‑management and consulting professionals have honed...
True Positive Weekly #159
The latest True Positive Weekly curates eight AI stories, from an AI‑driven breakthrough in mathematics to Google’s privacy‑costly Gemini defaults, and a Google‑led push to embed generative AI in workforce training. OpenAI’s new policy bans fantasy content like goblins, while...
Alibaba's Metis Agent Cuts Redundant AI Tool Calls From 98% to 2% — and Gets More Accurate Doing It
Alibaba researchers unveiled Hierarchical Decoupled Policy Optimization (HDPO), a reinforcement‑learning framework that teaches AI agents to balance task accuracy with execution efficiency. Using HDPO, their multimodal Metis agent slashed unnecessary tool invocations from 98% to just 2% while setting new...
-2.png)
How Is AI Voice Assistant Adoption Really Going in 2026? G2's AI Market Report Has the Answer
G2’s 2026 AI Voice Assistant Market Report shows that adoption is accelerating, with 88% of platforms reporting rapid growth over the past year. A striking 76% of users claim significant or transformational operational ROI, yet measuring that ROI ranks as...
Firefox Maker Torches Google for Building Prompt API Into Browser
Mozilla has publicly opposed Google’s Prompt API, a new browser feature that lets web pages run local AI models such as Gemini Nano directly in Chrome and Edge. The API requires roughly 22 GB of free disk space for the model,...
I Turned Down a Near-Million Dollar Job With OpenAI. Now My App Has 500,000 People On the Waitlist.
Former Stanford PhD candidate Div Garg turned down a near‑million‑dollar job at OpenAI to found AGI Inc., a startup building a voice‑driven AI assistant that runs on mobile devices. The company secured an $8 million seed round and has amassed about...
AI Twin Amplifies My Principles for Public Access
AI isn’t about replacing humans. It’s about compounding what we know and making it accessible at scale. Over time, I’ve learned that the most valuable thing I can do is capture and refine my principles on life, work, geopolitics, and...
Seven Things Every Medical Device Manufacturer Must Know Before Integrating AI
Artificial intelligence has moved from experimental to production‑ready in healthcare, enabling devices like predictive glucose monitors and AI‑driven oncology tools. For medical‑device makers, integrating AI is now a business imperative but also a complex regulatory undertaking. The article outlines seven...
Tech Industry Backs Anthropic In Copyright Battle With Music Publishers
Tech industry groups including NetChoice and the Computer & Communications Industry Association filed a friend‑of‑the‑court brief supporting Anthropic's claim that using song lyrics to train its Claude model is fair use. The brief argues that AI training involves non‑expressive, transformative...
Agentic Systems Add New Layer of AI Hallucination Risk in Legal Work
Law.com reports that emerging agentic AI systems—software that can act autonomously—are introducing a new class of hallucination risk in legal work. Unlike traditional generative models that require user prompts, agentic tools can initiate tasks, retrieve data, and draft contracts, increasing...
How Libraries Can Help the Public Engage with AI and Their Local Government
Public libraries are adding artificial‑intelligence programs to help residents move from awareness to practical use, positioning themselves as community tech hubs. Cities such as San Jose, Osceola County, and Frisco have opened AI learning centers, coding labs, and even a “library...
Marine Corps Turns to AI to Help ID Pay Delays, Boost Retention
The Marine Corps’ 4th Logistics Group has deployed an AI‑driven tool called PULSE Check to capture real‑time feedback on pay delays, training gaps, and other stressors affecting reservists. The four‑minute mobile survey generates statistically reliable data instantly, replacing the months‑long...
New Second Brain Update, Open Source Skills, and a Request: Show Us How You Use It
Hexact released a public GitHub repository containing five ready‑to‑use Claude skills that extend the Second Brain AI assistant, including a core permission‑first database access skill. The latest app version adds a no‑SQL database management tool and a global search bar...
VCs Fully Embrace AI for Operations
Venture capital firms are rapidly integrating AI agents into core operational workflows, from deal sourcing and due diligence to portfolio monitoring and internal reporting. A recent industry survey shows that over 70% of VC firms now use at least one...
AI Reveals Hidden Revenue, Not Just Headcount Cuts
If you're in a meeting right now talking about how AI will let you cut 40% of your team, you're playing the small game. IKEA just showed why. They deployed a chatbot named Billy to handle level 1 customer service. It...
AI Turns Menu Text Into Realistic Dish Images Instantly
karpathy just admitted that his own app got oneshotted and he thinks yours is next. he built menu gen. where you take a photo of a restaurant menu and it shows you pictures of what the food actually looks like (because 30-50% of...
ACM TechBrief: AI ‘Vibe Coding’ Could Reshape Software Development but Lacks Key Safeguards
The ACM Technology Policy Council’s new TechBrief warns that “vibe coding”—AI‑assisted software development driven by natural‑language prompts—can dramatically accelerate code creation but often bypasses essential engineering safeguards. While developers report higher productivity, the brief highlights rising security flaws, inconsistent testing,...
No AI Moat Yet; Runaway AI Remains Science Fiction
100%. There is no big AI lead. No AI company has demonstrated a sustainable moat. There's speculation that a frontier company will reach rapid self-improvement and run away from the crowd. This is, thus far, science fiction. There's zero sign...
AI Makes Fun Custom Songs, but Lacks Lasting Appeal
Had a “wow” moment with Suno: creating a hyper-personalized song for my kids, with the theme being our vacation adventures. Epic hit. At the same time, I cannot see myself using the app beyond this one-off neat party trick, nor listening...
C.H. Robinson Posts 15% EPS Rise, Leverages AI to Offset Spot‑Market Costs
C.H. Robinson reported a 15% year‑over‑year increase in adjusted earnings per share for Q1 2026 while keeping North American Surface Transportation (NAST) gross margin at 14.6%. The logistics firm credited a Lean AI strategy and disciplined volume mix for offsetting...
Balancing Trust, Flexibility, and Lock‑In Avoidance in 2026 Enterprise AI
The enterprise AI landscape in 2026 is shaped by three conflicting themes: trust, flexibility, and avoiding vendor lock-in. Navigating these competing demands will be key. #AI #EnterpriseTech https://t.co/qSVSGpL55L
AI Beats Doctors in Emergency and Complex Diagnosis Reasoning
An AI program performed better than human doctors on reasoning tasks such as making emergency room decisions and diagnosing complex cases, according to a new study published Thursday in @ScienceMagazine https://t.co/zAeOYW32I4 https://t.co/SGgO3KNfqM
VC-Backed CTOs Reveal How They're Leveraging AI
We have surveyed @2048vc portfolio CTOs on their usage of AI. Check out what they shared and comment on your usage that's not covered below.
Apple CEO Confirms New Siri Launch at WWDC June 8
Cook reiterates Siri coming this year. He also told CNBC that he's excited for the new Siri at WWDC. All eyes are on WWDC26 June 8th.
Press Gazette Analyzes DCN Amicus Brief on Agentic AI
Press Gazette continues to be all over the developments across AI, especially those impacting publishing. This is a really fair assessment of DCN's amicus brief filed overnight in the ninth circuit in Amazon v Perplexity - all about agentic AI...
Know When AI Isn't the Right Choice
When Not to Use #AI by Benjamin Laker @mitsmr Learn more: https://t.co/CVVXM6ikiZ #ArtificialIntelligence #MachineLearning #DL https://t.co/jF1sIYPcEb
Use AI to Create a Sustainable Digital Filing System
🗂️ End of the school year? 📁 AI can help you build a system for organizing digital resources that actually works. ⏰ Try 15 minutes every Thursday with a naming convention AI helped you create. https://t.co/GDB21nrz89
Selective LLM Regulation Is Performative, Drives Local Adoption
Regulating LLMs for biology but not every other potentially dangerous activity like turning on a gas stove in your home (unintentional fuel-air explosion device) is just performative knee-jerk and will just force people to use local LLMs. AGI soon but not...
AI Designs Life Using only 19 Amino Acids
Not something you'd see everyday—changing the alphabet of life. All of life organisms are are built from 20 amino acids. Now genAI is enabling life to be built with 19 amino acids, making isoleucine dispensable. @ScienceMagazine https://t.co/7CBn0Xhuxs https://t.co/tkxtCrFx9Y
AI Beats Human in F1 Sim Racing Challenge
This is so cool @brcullum built a machine learning model to race @dvassallo's #vibejam F1 sim and ran thousands of simulations to become the fastest Now we could have AIs competing against AIs in F1 races (like that IRL competition already does?)
Rigorous Solution Proposed for AI Unemployment Domino Effect
one of the better papers i've read (way outside of my expertise) about the domino effect of AI unemployment. i like how they try to find a *solution* in a rigorous way, as opposed to merely whining about the problem,...
Judge Bars Musk From Discussing AI Extinction Risks
Dear @elonmusk, If you still genuinely care about AI safety, you can’t let the Trump administration leave the AI industry almost entirely unregulated. You just can’t. - Gary
AI's Role in Misinformation and Deep‑Fake Threats
Twin pieces by @kakape in the new issue @ScienceMagazine on AI, used as a purveyor of misinformation and deep fakes, open-access https://t.co/lw22frYbKY https://t.co/HWm659MERp https://t.co/akkXh59eYB
AI Productivity Gains Modest Now, Will Surge with New Models
AI productivity gains: More modest than expected https://t.co/xW6Scx6ATH < checks out, so far. But once the agentic operating model really takes hold, team structures change, and better platforms improve the build-to-prod stages, these numbers will jump.