Get the latest AI news, understand why it matters, and learn how to apply it in your work — all in just 5 minutes a day. Join over 2,000,000+ subscribers.

Join Free
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
AI

The enterprise shift OpenAI saw coming

Zach Mink • 7 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. Two months ago, OpenAI's Fidji Simo told staff the company was in a "code red" over Anthropic's enterprise rise. New spend data says the leaderboard has actually flipped.

Ramp’s latest AI Index just showed Anthropic leading adoption among its paid business users for the first time, a 4x adoption surge since 2025 that may tell the story of why OpenAI has been drastically shifting its priorities throughout 2026.

P.S. — Our next live workshop, 'Finally Getting AI to Do Real Work', is today at 2 PM EST. Join for a breakdown of how AI actually works in 2026 and the context habits power users layer on top to get reliable output from any model. RSVP here.

In today’s AI rundown:

  • Anthropic seizes OpenAI's business AI lead

  • Amazon doubles down on Alexa+ for shopping

  • Create content with Claude Code and Higgsfield

  • Adaption automates AI training with AutoScientist

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

ANTHROPIC & OPENAI

🏢 Anthropic seizes OpenAI's business AI lead

Image source: Ramp

The Rundown: Fintech firm Ramp just published its latest AI Index, showing Anthropic taking the paid business-adoption lead from OpenAI for the first time, an enterprise surge that has quadrupled its usage over the past year while OpenAI has leveled off.

The details:

  • Ramp tracks corporate card and invoice payments from 50K+ U.S. businesses, making this a spend signal, not full market share.

  • Anthropic rose 3.8% in April to 34.4% of adoption, while OpenAI fell 2.9% to 32.3% as overall AI use continued to climb to 50.6%.

  • Claude Code anchored much of the swing, with Anthropic expanding from technical teams into finance, legal, and research workflows.

  • Ramp highlighted several risks facing Anthropic despite the trend, including recent Claude outages and increasing costs compared to OAI and open-source.

Why it matters: OpenAI is not suddenly cooked, with Ramp not tracking some large enterprise deals and ChatGPT remaining the bigger consumer brand. The yearly trend speaks louder, and was likely the same type of chart that caused OAI’s Fidji Simo-led pivot, which has become evident from recent Codex and other enterprise pushes.

TOGETHER WITH GOOGLE CLOUD

🤖 Build a Multi-Agent AI System

The Rundown: Scale your AI operations with the Gemini Enterprise Agent Platform, Google’s new evolution for building, governing, and optimizing AI agents. Using the open-source Agent Development Kit (ADK), this hands-on codelab teaches you to orchestrate specialized agents like Researchers and Judges into self-correcting workflows that solve complex problems at enterprise scale.

In this codelab, you will:

  • Construct self-correcting AI feedback loops.

  • Orchestrate task hand-offs between specialized agents.

  • Deploy production-ready workflows to Google Cloud Run.

Build and deploy your first Multi-Agent system with Google.

AMAZON

🛒 Amazon doubles down on Alexa+ for shopping

Image source: Amazon

The Rundown: Amazon folded its standalone shopping chatbot ‘Rufus’ inside "Alexa for Shopping", a new agent that takes over Amazon search and follows shoppers across devices with a shared memory of purchases, preferences, and prior chats.

The details:

  • Rufus drew 300M+ users in 2025 while still in beta, with its product knowledge and shopping history now feeding Alexa for Shopping answers.

  • Amazon says the new assistant draws on catalog data, reviews, delivery timing, past purchases, and Alexa conversations for information.

  • Alexa can now field questions in the search bar, run side-by-side comparisons, track pricing, and Auto-Buy items when prices hit a target.

  • A new Buy for Me handles checkouts on non-Amazon stores, with Scheduled Actions able to automatically restock products on a cadence.

Why it matters: RIP to Rufus, but consolidating under one Alexa AI brand feels like a better play, and the sheer amount of customer history Amazon has gives its agent quite the moat to work with. But with a consumer base that is already moving to other AI platforms for their agentic shopping needs, will Alexa play nice and integrate with them?

AI TRAINING

📸 Create content with Claude Code and Higgsfield

The Rundown: In this guide, you will learn how to connect Higgsfield to Claude Code with the Higgsfield CLI, then use Claude Code to send one image prompt to several AI image models at once.

Step-by-step:

  1. Create a new project folder, install Higgsfield CLI (npm install -g @higgsfield/cli), authenticate (higgsfield auth login), and add Higgsfield skill (npx skills add higgsfield-ai/skills) for Claude Code

  2. Open Claude Code in the same folder and ask it to inspect higgsfield.ai/cli, verify the installation, and list available image models

  3. Give Claude Code an image prompt and tell it to run it across six models, save outputs into a higgsfield-model-test folder, and generate a comparison.md file with notes for each result

  4. Pick the best direction, then ask Claude Code to refine the winning prompt or run one more comparison

Pro tip: If setup gets confusing, ask Claude Code to check Node/npm and give a walkthrough. Higgsfield also supports video, so you can also try this for short clips.

PRESENTED BY STRIPE

💰 Pricing AI: Lessons from leading AI companies

The Rundown: AI companies are hitting revenue milestones faster than ever—but figuring out how to monetize AI products is still one of the hardest problems founders face. Stripe interviewed teams at Anthropic, Vercel, Clay, and more to build a five-step pricing framework for AI products.

Download the guide to learn best practices for how to:

  • Pick the right pricing model

  • Prevent billing surprises

  • Refine your pricing over time

Get the framework. 

ADAPTION

🔬 Adaption automates AI training with AutoScientist

Image source: Adaption

The Rundown: Adaption, the AI startup from ex-Cohere VP of Research Sara Hooker, just introduced AutoScientist, a new system that automatically customizes AI models for specific jobs by adjusting both what the model learns from and how it learns.

The details:

  • AutoScientist tests different training data and settings, then iterates until the model meets the user's goal.

  • In internal tests, AutoScientist outperformed its own expert-tuned models by 35% on average, with success rates jumping from 48% to 64%.

  • Results held across multiple AI models, a wide range of dataset sizes, and 8 varied industries, including finance, legal, and medical domains.

  • Adaptive’s initial Adaptive Data release in February aimed at increasing the quality of datasets, with Autoscientist now moving toward customizing models.

Why it matters: A few thousand people in the world know how to properly train and fine-tune a frontier model, and nearly all of them work at the same handful of labs. If a tool like Autoscientist can start to automate that expertise, models customized for individual businesses and use cases may become a lot more practical to create.

QUICK HITS

🛠️ Trending AI Tools

📰 Everything else in AI today

Nvidia became the first company to hit a $5.5T market cap, coming as CEO Jensen Huang arrived in China to join U.S. President Donald Trump in meetings with Xi Jinping.

Sam Altman testified in the Elon Musk vs. OpenAI legal battle that Musk’s “specific plans on safety” made him worry, and proposed passing the company to his children.

David Silver’s Ineffable Intelligence and Nvidia announced a partnership to build training pipelines for RL agents, with early work targeting Nvidia's Vera Rubin hardware.

Microsoft introduced MDASH, an AI security harness that chains 100+ specialized agents to hunt for software bugs, with the system catching 16 flaws across Windows.

UK's AI Safety Institute said that AI’s ability to complete cyberattacks is doubling every few months, with Mythos Preview and GPT-5.5 finishing its simulated breaches.

COMMUNITY

🤝 Community AI workflows

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Rod R. in Elizabethtown, KY:

"I am an avid cyclist. I turned 61 years old this year, but I am still participating in multi-day cycling events. I needed advice on the type of riding I should be doing to prepare, and I thought I could benefit from a coach, but I can’t afford one. Then it occurred to me to try ChatGPT.

That has worked out tremendously! We had a “conversation” about my goals, medical history, current fitness, event dates, etc. I have been following a training plan for about 6 weeks now.

ChatGPT continues to coach me as I report on each ride, helps me restructure the plan when life requires me to modify it, and also remembers things we have already discussed. I love how this has helped me to prepare, and I have peace of mind from knowing what I should be doing each day. Today is a rest day! 😊"

How do you use AI? Tell us here.

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — the humans behind The Rundown

AI

Android enters its Gemini Intelligence era

Zach Mink • 7 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. Google’s headlining I/O event isn’t until next week, but Android just gave a pretty big performance for an opening act.

The company just introduced a new line of AI-native Googlebooks, a new Gemini Intelligence system across devices, a Gemini-infused mouse cursor, and more — bringing AI into its device ecosystem in a more unified way than ever before.

In today’s AI rundown:

  • New Googlebooks, Gemini Intelligence for Android

  • Google circles SpaceX for orbital AI compute

  • Turn Claude Code into your Wall Street analyst

  • Amazon's AI scoreboard warps work incentives

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

GOOGLE

💻 New Googlebooks, Gemini Intelligence for Android

Image source: Google

The Rundown: Google just rolled out major new Gemini integrations and hardware at its Android Show event, including a new line of AI-native Googlebook laptops, a Gemini Intelligence system for devices, an AI-infused mouse cursor interface, and more.

The details:

  • Googlebooks ship this fall as Gemini-native laptops built with Dell, HP, Lenovo, Acer, and Asus, featuring a 'Magic Pointer' AI cursor shown in a new demo.

  • These new laptops will run Android phone apps and files, blending ChromeOS, Android, Google Play, and Gemini.

  • Gemini Intelligence acts as Android’s cross-device AI platform, able to carry out agentic tasks within apps and operate with on-screen context.

  • Other releases include a Create My Widget tool, a Rambler dictation tool that strips filler words, Gemini auto-browse in Chrome on-device, and more.

Why it matters: I/O isn’t until next week, but this was a pretty big appetizer. While the world awaits Apple’s Siri AI revival, Gemini is being woven directly into Android instead of as another bolted-on feature. An ‘intelligence system’ across devices is a clear path to making AI actually useful, and Google might be the first one to actually crack it.

TOGETHER WITH OPTIMIZELY

🧪 Where AI meets the future of experimentation

The Rundown: Test + Learn 2026, hosted by Optimizely, is a free two-hour virtual event on June 17 built for digital leaders putting AI to work in their experimentation programs. Expect honest takes, live agent demos, and real frameworks teams are using today.

At the event, you'll:

  • Hear career-shaping AI advice from Elena Verna and top minds in the field

  • See 5 live AI agent demos covering everything from ideation to reporting

  • Learn how Salesforce, BBC, ASOS, and more run more tests without extra headcount

  • Get exclusive content and an invite to free AI certificate training

Register now.

GOOGLE & SPACEX

🛰️ Google circles SpaceX for orbital AI compute

Image source: Images 2.0 / The Rundown

The Rundown: Google is reportedly exploring a rocket-launch deal with SpaceX for orbital data centers, putting two already-linked companies on the same side of a moonshot that could also become a future AI infrastructure rivalry.

The details:

  • Google has held a 6.1% stake in SpaceX since investing $900M in 2015, and its VP Don Harrison occupies a board seat at the company.

  • Google’s own moonshot Project Suncatcher is aiming to launch prototype Google satellites by 2027, with Planet Labs helping build the first hardware.

  • Anthropic finalized a compute deal with SpaceX last week, with Anthropic also saying it “expressed interest… in multiple GW of orbital AI compute capacity”.

  • SpaceX has filed for approval for up to 1M satellites, making orbital compute a major piece of its pre-IPO pitch to investors.

  • OAI's Altman called the concept of orbital compute 'ridiculous' at an event in New Delhi, saying it won't 'matter at scale this decade.

Why it matters: The deal makes a lot of sense for both sides — Google gets launch capacity for Suncatcher without building its own rockets, and SpaceX gets a marquee customer to validate its orbital pitch before its IPO. Despite Altman’s sentiment, betting against ideas from both Google and Elon Musk hasn’t been a great move historically.

AI TRAINING

📈 Turn Claude Code into your Wall Street analyst

The Rundown: In this guide, you will learn how to add Anthropic's financial services plugin marketplace to Claude Code, install market research skills, and use them to create market research reports, equity analysis, earnings reviews, and Excel sheets.

Step-by-step:

  1. Open a terminal and start Claude Code. Once it’s running, type: /marketplace and hit enter on the plugin marketplace option to open the plugin manager

  2. Go to the Marketplaces tab in the plugin manager, select Add Marketplace, and paste this GitHub link to add the new claude-for-financial-services marketplace

  3. Open the new marketplace and browse plugins. For us, market-researcher, earnings-reviewer, equity-research, and financial-analysis did well

  4. Install the skills you want, then use them inside Claude Code to create sourced reports, compare companies, review earnings, or financial analysis outputs

Pro tip: Treat this as research support, not financial advice. Use public information, ask Claude to cite sources, and verify anything important before making decisions.

PRESENTED BY ASAPP

♻️ Five CX agents, one coordinated system

The Rundown: Most enterprises automate containment and call it done. ASAPP's five new coordinated agents — Discovery, Developer, Simulation, Optimization, and Insights — take you from finding what to automate to building, testing, running, and improving it all in one governed platform with human oversight built in.

ASAPP’s agents can:

  • Spot highest-ROI automation opportunities and ship CX workflows without the engineering wait

  • Test new experiences before launch and tune live performance continuously

  • Turn every customer interaction into intelligence the business can act on

Meet ASAPP’s new agents.

AMAZON

🧮 Amazon's AI scoreboard warps work incentives

Image source: Images 2.0 / The Rundown

The Rundown: Amazon's internal AI ‘tokenmaxxing’ push has resulted in employees gaming its MeshClaw agent to burn extra tokens, with staff telling FT that adoption and public consumption metrics have turned token usage into an office competition.

The details:

  • The company had set an internal goal for over 80% of developers to use AI weekly, and started tracking model and token usage via staff rankings this year.

  • MeshClaw, a tool built by Amazon staffers, lets users create AI agents with access to deploy code, sort emails, and operate across company software.

  • Amazon staff told FT the pressure to adopt AI is causing “perverse incentives”, with employees burning tokens on unnecessary tasks to raise their numbers.

  • Amazon says token stats are not performance-review inputs, but recently pulled back usage number visibility to individual employees and their managers.

Why it matters: The quality > quantity adage feels like an important lesson for the rise of tokenmaxxing seen within companies like Amazon and Meta. A token counter can prove AI was used, but it can’t prove the work got better — and if companies reward usage instead of outcomes, employees will only optimize for the scoreboard.

QUICK HITS

🛠️ Trending AI Tools

  • 🧠 Pave -  Skip prototype purgatory. Build custom AI apps your business can actually run. No tools, setup, or cost*

  • 🌌 Krea 2 - Krea's image AI with style transfer, moodboard-based generation

  • 🎨 Step Image Edit 2 - StepFun’s new lightweight image editing model

  • 🗣️ GPT-Realtime-2 - Voice AI that thinks, call tools, maintains flow in live calls

*Sponsored Listing

📰 Everything else in AI today

Google’s Isomorphic Labs announced $2.1B in funding for its drug discovery AI, with Demis Hassabis saying “the No. 1 application of AI should be to improve human health.”

Krea introduced Krea 2, the company’s first proprietary image model specifically designed for aesthetic range, with features like style transfer and moodboard tools.

A hacker planted data-stealing code called ‘Mini Shai-Hulud’ inside 42 of its open-source agentic npm packages, with the wide-scale attack impacting several AI tools.

Meta employees reportedly organized a protest against the company’s mouse-tracking software being used to train AI and create agents.

Rivian rolled out an AI assistant across its EVs that controls car hardware, chains agentic tasks, and more via a steering wheel button or “Hey Rivian” command.

COMMUNITY

🤝 Community AI workflows

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Corey S. in Ft. Lauderdale, FL:

"Six months ago, my little brother passed away unexpectedly. On the night before his funeral, we realized that the hundreds of people attending likely held a treasure trove of photos we had never seen—snapshots of his life from childhood to adulthood.

On the morning of the service, I quickly vibe-coded and deployed a simple site that allowed guests to scan a QR code and upload photos directly from their phones.

As people arrived, a live wall of memories began to grow, giving everyone a chance to reflect on his life while they waited. Afterward, I was able to download every single one of those photos and send them to my mother."

How do you use AI? Tell us here.

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — the humans behind The Rundown

Tech

Venmo finally kills its most criticized feature

Jennifer Mossalgue • 5 minutes

Read Online | Sign Up | Advertise

Good morning, tech enthusiasts. Venmo is finally overhauling its payments app. After years of criticism that its public-by-default social feed exposed users’ routines, relationships, and spending habits to anyone paying attention, the PayPal-owned platform is making privacy a core part of the experience.

It’s the app’s biggest redesign since its inception, and a long-overdue reset for a company that turned oversharing into a growth strategy.

In today’s tech rundown:

  • Venmo rolls out privacy-focused redesign

  • Lime, the Uber-backed scooter giant, files for IPO

  • California sues Meta over scam ad profits

  • Whoop adds on-demand doctors to its data

  • Quick hits on other tech news

LATEST DEVELOPMENTS

VENMO

💵 Venmo rolls out privacy-focused redesign

Image source: PayPal

The Rundown: PayPal-owned Venmo is rolling out its biggest app redesign since its inception in 2009, and it’s finally defaulting new users to friends-only payments instead of broadcasting their transactions to the world.

The details:

  • New users’ posts will now be visible to friends only by default, rather than the public; the option to restrict further to “just me” is also available.

  • An updated send screen will display each transaction’s privacy status before it goes out, so users know what they’re sharing before they hit send.

  • The redesigned feed adds emoji reactions and quick actions like “Pay Again” and “Say Thanks,” plus a “Give a Shoutout” button for local businesses.

  • The rollout begins this week on iOS and Android, with the full redesign expected to reach all users by fall.

Why it matters: Venmo is finally walking back its public-by-default social feed, which critics have long warned could expose users’ daily routines and relationships. Re-centering the app on friends-only and private payments shows that basic data protection is starting to outweigh growth hacks built on sharing all the details.

TOGETHER WITH ASTROCADE

🎮 Where AI meets social gaming

The Rundown: Astrocade is revolutionizing user-generated gaming. Backed by $56M from Sequoia, it’s the ultimate platform to seamlessly scroll, play, and instantly build your own games for a massive global community.

With Astrocade, you get:

  • Free AI editor to generate or import assets.

  • Create viral games and earn real money.

  • Build high-quality, complex games using powerful AI.

Play and create now.

LIME

🛴 Lime, the Uber-backed scooter giant, files for IPO

Image source: Lime

The Rundown: Lime, the neon-green scooter and e-bike operator that has been flirting with public markets since 2020, has finally pulled the trigger — filing its S-1 with the SEC and targeting a $2B valuation.

The details:

  • Lime grew revenue from $521M in 2023 to $686.6M in 2024 and $886.7M in 2025, though the company posted net losses of $59.3M last year.

  • On the brighter side: Lime has generated free cash flow three years running, hitting $104M in 2025, nearly double the prior year.

  • Lime is still carrying around $1B in current liabilities, with roughly $846M due within 12 months and $675.8M owed by year’s end.

  • Uber remains deeply embedded — exclusive in-app integration accounts for roughly 14–16% of Lime’s total revenue.

Why it matters: Lime is one of the last major micromobility players still standing after Bird went bankrupt in 2023, and its filing marks a pivotal moment for a sector riddled with financial wreckage. If Lime can’t persuade public investors that shared scooters are a real business, the category may not get another shot.

META

⚖️ California sues Meta over scam ad profits

Image source: Images 2.0 / The Rundown

The Rundown: Santa Clara County in California sued Meta, accusing the company of knowingly profiting from a flood of scam ads on Facebook and Instagram while publicly overstating its enforcement efforts.

The details:

  • California’s Santa Clara County filed suit against Meta, alleging violations of state false advertising and unfair competition laws.

  • The complaint, drawing on leaked internal documents reported by Reuters, alleges Meta generated as much as $7B annually from “high-risk” advertisers.

  • Prosecutors say the company’s ad‑targeting tools and AI systems helped scammers micro‑target victims and rapidly scale their schemes.

  • The county claims Meta publicly overstated the strength of its ad review and anti‑fraud safeguards while failing to curb obvious scam activity.

Why it matters: The suit is among the most aggressive legal actions yet to allege a major platform’s ad business structurally helped fraud rather than simply failing to police it. If Santa Clara County can prove that Meta capped enforcement to protect revenue, the legal and regulatory exposure for the industry could be significant.

WHOOP

🩺 Whoop adds on-demand doctors to its data

Image source: Whoop

The Rundown: Whoop is turning its screenless fitness band into a quasi-virtual clinic, adding on-demand video consultations with licensed clinicians just as Google readies a Whoop-style Fitbit Air to chase the same always-on-health crowd.

The details:

  • Live video consultations with U.S.-based licensed clinicians roll out in-app this summer as a paid add-on to a standard Whoop membership.

  • Each session is built on months of continuous biometric data and, when available, bloodwork and medical history.

  • A new partnership with health records keeper HealthEx enables EHR syncing, letting clinicians pull up info and procedures directly within the app.

  • Whoop, which has over 2.5M members globally, closed a $575M funding round in March that pushed its valuation to $10.1B.

Why it matters: Whoop is betting that months of continuous biometric data — not a 15-minute visit — becomes the foundation of primary care. With Google’s Fitbit Air entering the screenless band market at $99, the fight looks to be less about hardware and more about what you can do with all of that longitudinal health data.

QUICK HITS

📰 Everything else in tech today

Amazon is expanding its “Amazon Now” service — which delivers groceries, household essentials, and other items in 30 minutes or less — to dozens of U.S. cities.

Trump invited Elon Musk, Tim Cook, Larry Fink, and a group of other top U.S. CEOs to join his trip to China this week for a summit with President Xi Jinping.

The FCC will now let already‑approved foreign-made routers and drones keep receiving updates until 2029, slightly easing its earlier ban on such devices.

China plans to expand its Tiangong space station from a T-shaped outpost into a larger six‑module complex with international astronaut access, as the ISS retires.

A QTS data center in Georgia reportedly consumed nearly 30M gallons of unmetered water during a drought due to two unsupervised connections.

TikTok is rolling out a £3.99-per-month ad-free subscription in the U.K., removing ads and ad-targeting data use, likely in response to local GDPR privacy rules.

Texas is suing Netflix, accusing it of secretly harvesting and monetizing users’ and children’s data in violation of Texas consumer protection laws.

Astronomers used James Webb Space Telescope data from the COSMOS-Web survey to create the most detailed map yet of the universe’s cosmic web.

Prosus aims to generate about $3.6B in annual revenue from Just Eat within a year as it pushes to streamline and grow its European food-delivery operations.

COMMUNITY

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — The Rundown’s editorial team

AI

Mira Murati's TML upends how humans work with AI

Zach Mink • 7 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. Both Mira Murati's Thinking Machines and Ilya Sutskever's SSI have spent the post-OpenAI era mostly out of view, making every public reveal feel that much bigger.

Murati's lab just broke the silence with 'interaction models,' a new type of AI built for real-time collaboration across voice, video, and text — in a direct counter to the agentic-first direction the rest of the field is racing toward.

In today’s AI rundown:

  • TML’s new interaction models for real-time AI

  • Google traces software attack back to AI

  • Build a YouTube research bot in 15 minutes

  • Anthropic fixes Claude's blackmail problems

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

THINKING MACHINES LAB

🗣️ TML’s new interaction models for real-time AI

Image source: Thinking Machines Lab

The Rundown: Thinking Machines Lab (TML) just introduced a research preview of interaction models, a new kind of AI system built to collaborate live across voice, video, and text — letting users talk, show, interrupt, and steer while the system keeps working.

The details:

  • The model takes in voice, video, and text in 200ms chunks, perceiving and responding in a streaming loop without the turn-taking pauses of other rivals.

  • A second background model handles slower reasoning, searches, and tool work, allowing the live model to keep talking and interacting with the user.

  • The system can also react to visual changes, count reps, translate live speech, and speak up at timed moments instead of waiting.

  • CEO Mira Murati said TML is focused on advancing human-AI collaboration, and that “the way we work with AI matters as much as how smart it is.”

Why it matters: Murati's TML has been fairly quiet since its inception, but interaction models are one of the lab’s first big differentiators: models designed around how people naturally work together, not how long an agent can run solo. Whether it carves out its own market or gets absorbed by a frontier lab's next update is the question now.

TOGETHER WITH YOU.COM

🧠 What’s the point of an LLM if it hallucinates?

The Rundown: It happens—LLMs hallucinate. Grounding your LLM, however, can help dramatically improve accuracy. In this guide, You.com explains what AI grounding is and how organizations can implement it to achieve more reliable outputs.

The playbook covers:

  • A three-part approach that outperforms RAG alone

  • Why grounding isn't set-and-forget, and how to build audit trails

  • The open vs. closed platform trade-off (and what it means for your next model switch)

Get the guide.

GOOGLE

🔒 Google traces software attack back to AI

Image source: Google

The Rundown: Google's Threat Intelligence Group confirmed the first known case of hackers using AI to discover and write a zero-day software security flaw, catching them before they could break past login protections on a widely-used web management tool.

The details:

  • The hack was intended to allow the user to get around two-factor authorization on the affected app, with Google working with the company to stop the attack.

  • Google pointed to unusually polished attack code, long explainer notes, and a made-up severity score as clues that the exploit was written with an AI.

  • GTIG's John Hultquist called the find "the tip of the iceberg," with Anthropic's Rob Bair warning cybersecurity defenders' lead is "months, not years.”

  • GTIG detailed other hacks, including software that lets AI remotely control a device, and AI-assisted malicious prompts and code from N. Korea and Russia.

Why it matters: We’ve already started to see what Anthropic’s Mythos can do on the cybersecurity front, but attackers aren’t too far off from having similar power. Even with careful rollouts, the next step up the release ladder is about to open the door to some serious security issues that will cause chaos for the many systems not ready for it.

AI TRAINING

📺 Build a YouTube research bot in 15 minutes

The Rundown: In this guide, you will learn how to build a Gumloop agent that tracks YouTube channels or search topics, reads transcripts, and turns the useful videos into a ranked research brief.

Step-by-step:

  1. Go to Gumloop agent builder, create an agent named YouTube Scout, and enable YouTube and Google Sheets in the right-hand section under "Apps"

  2. Prompt: Build me a YouTube scout for (niche). Check (channels/queries), find videos from the last (hours/days), read the transcript, and return a brief with title, link, 3-5 takeaways, why it matters, follow-up ideas, usefulness score, and a "what changed" summary. Track topics and videos in a Google Sheet

  3. Start small: one niche, a few trusted channels, one or two searches, and a 24-48 hour lookback window. The tighter the scout's beat, the better the brief

  4. Run the agent, then review the Sheet it creates. Make sure each result has a source link, concrete takeaways, and a usefulness score

Pro tip: Dial in the signal score early. If the agent calls a mediocre video an 8, tell them why that should be a 5. You can also add a User Signal Score column for future runs.

PRESENTED BY TELY AI

💬 Market leaders get leads from ChatGPT and Google

The Rundown: Your buyers are asking AI questions — and AI is answering with your competitors, not you. Tely makes AI like ChatGPT, Google, and Claude recommend your business instead.

With Tely AI, you can:

  • Get recommended in ChatGPT, Google, Perplexity, and Claude in as little as 1 week

  • Fully hands-off: no writers, no agencies, no managing content

  • Costs less than hiring freelancers or maintaining a marketing team

  • Ideal for niche industries where expertise matters

Get leads from Google and ChatGPT on autopilot.

AI RESEARCH

🐍 Anthropic fixes Claude's blackmail problems

Image source: Anthropic

The Rundown: Anthropic published a study detailing how it fixed Claude's previously seen blackmail behavior, highlighting the need to teach the model “why” and tracing the problem to internet fiction that depicts AI as power-seeking and self-preserving.

The details:

  • Earlier tests put Claude models in fictional workplace situations, with older systems resorting to blackmail and threats to avoid shutdown.

  • Having Claude reason through ethical choices, not just copy the safe action, cut blackmail rates from 96% in Opus 4 to nearly 0% for every model after.

  • Fictional stories of well-behaved AI and constitution-based documents also helped reduce bad behavior by more than 3x.

  • Just 3M tokens of ethical reasoning data matched 85M tokens of behavioral examples, a 28x efficiency gain that held up in deeper training.

Why it matters: AI is still far from an exact science, and eliminating blackmail via essentially positive AI stories and constitution docs is another one of the many strange training quirks. A small dataset of ethical fiction outperforming 28x the behavioral data shows how much of alignment is still guesswork, even when the guesses work.

QUICK HITS

🛠️ Trending AI Tools

  • 🤖 Slackbot - Your AI assistant that searches, summarizes, and automates work right inside Slack*

  • ❤️ Lovable Aesthetics - Vibe coding with more control over layout, typography

  • ⚙️ Parallel Agents - Run up to 10 parallel computer-use agents in Replit

  • ☀️ Daybreak - OpenAI’s new Codex-driven cybersecurity product

*Sponsored Listing

📰 Everything else in AI today

OpenAI launched “The Deployment Company”, a $14B business to embed engineers inside enterprises to deploy its AI, also acquiring AI consulting firm Tomoro.

SoftBank’s Masayoshi Son is reportedly in talks for a $100B AI investment into France, with plans to build out new data centers in the country.

Anthropic reportedly signed a 7-year, $1.8B cloud infrastructure deal with Akamai, adding another compute avenue to power its Claude models.

China’s Kuaishou Technology is reportedly planning to turn its Kling AI video branch into its own company, with a projected valuation of $20B and plans to IPO in 2027.

Former OpenAI Chief Scientist Ilya Sutskever testified in the Elon Musk vs. OpenAI lawsuit, revealing his current shares of the company total nearly $7B.

COMMUNITY

🤝 Community AI workflows

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Sasha M. in Cape Coral, FL:

"I have a family of 5, and planning what we have for dinner was a nightmare. I have a Trello board full of hundreds of recipes that I use to plan our meals, and then I would place a grocery delivery order online. The whole process would take up to an hour.

I built a Claude plugin that includes multiple skills to help plan meals and order groceries. I have it on a schedule to run once a week. First, it asks me for details about the week: our schedule, any days that I'll have fewer than 5 people eating, etc. Using an MCP to Trello, the first Claude Skill picks out 7 recipes and presents them to me.

Once I've approved the meal plan, Claude then creates an ingredients list that I check off anything I already have in my fridge/pantry. The plugin then runs a skill that goes to my grocery store website and adds all the ingredients to my cart. All I have to do is check the cart and click ‘Order.’"

How do you use AI? Tell us here.

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — the humans behind The Rundown

Robotics

Figure's robots make a bed together

Jennifer Mossalgue • 5 minutes

Read Online | Sign Up | Advertise

Good morning, robotics enthusiasts. Figure just dropped a two-minute demo of its F.03 humanoids tag-teaming a full bedroom reset — opening doors, hanging clothes, and making a neatly smoothed bed.

The twist is that the robots share a single learned neural network, coordinating through visual cues alone, no explicit communication or central planner. Everyone is racing to build better robot brains; Figure just showed what happens when two bodies share one.

In today’s robotics rundown:

  • Figure’s humanoids make a bed together

  • South Korea’s army turns to Boston Dynamics

  • Uber’s robotaxi partner has a crash problem

  • Rocket Lab buys robot arm that went to Mars

  • Quick hits on other robotics news

LATEST DEVELOPMENTS

FIGURE AI

🛏️ Figure’s humanoids make a bed together

Image source: Figure AI

The Rundown: Figure AI just dropped a two‑minute demo of its Helix‑02–powered humanoids tag‑teaming a full bedroom reset — opening doors, hanging clothes, and making a neatly smoothed bed “better than most humans,” says CEO Brett Adock.

The details:

  • Figure calls it the first demo of a single learned neural network driving multi-humanoid collaborative locomanipulation, directly from pixels to actions.

  • The F.03 robots coordinate purely through a shared AI policy and visual cues, with no explicit communication or central planner directing their moves.

  • In the video, they open doors, hang clothes, clear surfaces, and then team up to lift, spread, and smooth a comforter to hotel‑room neatness.

  • The F.03’s wireless foot-charging dock lets it step onto a pad and recharge at 2 kW, no outlet-hunting or human intervention required.

Why it matters: While many multi-robot systems depend on explicit communication or a centralized planner, Figure shows emergent coordination — a significant architectural bet on scalability. If it holds outside staged demos, it’s the kind of foundation that makes a capable robot a deployable one.

HYUNDAI

🪖 South Korea’s army turns to Boston Dynamics

Image source: Images 2.0 / The Rundown

The Rundown: South Korea is reportedly looking to deploy Hyundai's robotics arsenal — including Boston Dynamics’ Spot — to shore up military ranks as plummeting birthrates drain the conscription pool.

The details:

  • South Korea’s military is in talks with Hyundai to deploy robots across surveillance, reconnaissance, and logistics as troop numbers decline.

  • Potential systems include the Spot robot dog, the wheeled MobED platform, and X-ble Shoulder exoskeletons for load-bearing and mobility assistance.

  • Active-duty troop numbers have dropped from 650K in 2020 to 450K today, and the country has one of the world’s lowest birthrates.

  • Military planners are shifting toward unmanned systems to handle tasks from perimeter patrols to supply runs, fundamentally reshaping front-line operations.

Why it matters: South Korea's troop shortage is already here, and robots are looking like one of the most viable stopgaps on the table. But deploying unproven systems near the DMZ, where a sensor failure doesn’t just stop a factory line, certainly raises questions about how far militaries should go in automating their military ops.

AVRIDE

🚖 Uber’s robotaxi partner has a crash problem

Image source: Avride

The Rundown: U.S. auto safety regulators have opened a formal investigation into Uber partner Avride after 16 crashes involving its self-driving robotaxis occurred in Dallas over the past four months.

The details:

  • U.S. auto safety regulators have opened a formal probe into Avride, a self-driving startup that operates robotaxis for Uber in Dallas.

  • The investigation was triggered by 16 crashes in which Avride vehicles were in autonomous mode with a human safety driver behind the wheel.

  • Reported incidents include vehicles changing lanes into moving traffic, hitting stationary objects, and failing to respond appropriately to cars ahead.

  • Regulators are scrutinizing whether Avride’s software is overly aggressive and whether its safety practices are adequate for operating on public roads.

Why it matters: The probe cuts to the core of Uber’s strategy: outsourcing autonomy to startups rather than building it in-house. If regulators find systemic flaws in Avride’s software or safety culture, expect a broader chill on robotaxi deployments, and fresh ammunition for skeptics who say AI drivers still aren’t ready for real streets.

ROCKET LAB

🚀 Rocket Lab buys robot arm that went to Mars

Image source: Rocket Lab

The Rundown: Rocket Lab just signed its largest launch contract to date — and revealed that it’s acquiring Motiv Space Systems, a Pasadena robotics firm whose hardware operated on the Perseverance Mars rover and NASA’s CADRE lunar rovers.

The details:

  • Motiv’s portfolio covers multi-degree-of-freedom robotic arms, actuators, and drive electronics purpose-built for deep-space conditions.

  • Rocket Lab will fold Motiv into a new division called Rocket Lab Robotics, absorbing its 50-person engineering team and Pasadena manufacturing facility.

  • The deal targets a specific pain point: solar array drive assemblies and other precision mechanisms that are expensive and hard to source.

  • Rocket Lab says the capability will extend into on-orbit and surface operations, positioning it for commercial Mars Sample Return and national security work.

Why it matters: Most space companies source their robotics from contractors and live with the bottlenecks. Rocket Lab is closing that gap, fusing launch, satellites, and robotics into a single in-house stack. By buying Motiv’s robots, it’s betting that owning the manipulators for on-orbit work matters as much as the rockets that get them there.

QUICK HITS

📰 Everything else in robotics today

Hyundai’s Boston Dynamics unit released a video of its production-ready Atlas humanoid performing complex gymnastics.

Drone startup Helsing is reportedly lining up a new $1.2B funding round that would lift its valuation to about $18B, as investors pour money into defense tech companies.

South Korean robotics startup RLWRLD, backed by LG Electronics, reportedly released RLDX-1, a foundation AI model designed for five-finger robotic hands.

Austin-based Allen Control Systems, an anti-drone AI startup that makes autonomous weapons stations, is in talks to raise $200M at a $2B valuation.

Bloomberg’s Oleg Matsnev argues that today’s viral humanoids are still clunky, limited machines that fall far short of their AI-hype billing.

Self-driving truck startup Kodiak AI reportedly raised $100M by selling shares at a steep 29% discount, causing its stock to plummet 37% in after-hours trading.

Croatian startup Verne launched what it says is Europe’s first commercial robotaxi service in Zagreb in partnership with Uber, charging a flat fee of €1.99.

Nanoleaf, the smart lighting company, is pivoting to a new product lineup focused on robots, AI, and red-light therapy wellness devices, The Verge reports.

China unveiled a 220 lb., four-wheeled robot with dual robotic arms for its upcoming Chang’e-8 lunar mission, designed to function as a “Moon mechanic.”

Hugging Face launched an agentic toolkit for Reachy Mini that lets users without coding skills create custom robot apps by describing tasks in natural language.

COMMUNITY

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — The Rundown’s editorial team

AI

Google DeepMind’s powerful AI co-mathematician

Zach Mink • 6 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. Google DeepMind just took AI’s coding strategy and applied it to math: don't ask a model for the answer, give a team of agents the workspace.

The company’s AI co-mathematician just scored a new high on a benchmark built to stump AI for decades, with one professor even cracking an unsolved problem using a strategy buried inside a proof the system's own reviewers had rejected.

In today’s AI rundown:

  • Google DeepMind’s AI co-mathematician

  • The Rundown Roundtable: Our AI use cases

  • Automate any manual task with Codex

  • AI finds 100+ new exoplanets from NASA data

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

GOOGLE DEEPMIND

🧮 Google DeepMind’s AI co-mathematician

Image source: Pushmeet Kohli (@pushmeet on X)

The Rundown: Google DeepMind just published a paper on its AI co-mathematician, an agentic system based on Gemini 3.1 built to help mathematicians tackle unsolved problems — setting a new high on a benchmark of research-level math problems.

The details:

  • DeepMind modeled the tool after AI coding environments like Claude Code, bringing agent teams and built-in review cycles to math research.

  • A coordinator agent breaks research into parallel workstreams, each with sub-agents that write code, search literature, and attempt proofs.

  • Oxford’s Marc Lackenby resolved an open problem in the Kourovka Notebook after spotting a 'really, really clever proof strategy' inside a rejected output.

  • On Epoch AI's FrontierMath Tier 4, the system topped the leaderboard at 48% and more than doubled Gemini 3.1 Pro's 19% raw score.

Why it matters: AI has already led to a surge in mathematics discoveries with the advances in frontier models, and similar to coding, agentic pipelines are now enabling AI systems to push even further. But as Lackenby’s discovery shows, the future is still bright for AI that enables top minds to accelerate their work, not replace it.

TOGETHER WITH GOOGLE FOR STARTUPS

📚 Master generative media for startups

The Rundown: Google for Startups' Future of AI report is your essential guide to understanding how generative media is reshaping product development, offering founders strategic insights to build smarter, scale faster, and stay ahead of the AI curve.

Inside the report, you’ll discover:

  • How to leverage digital twinning at scale.

  • Strategic insights for AI product differentiation.

  • Expert perspectives on the generative landscape.

Download the report today.

THE RUNDOWN ROUNDTABLE

💡 The Rundown Roundtable: Our AI use cases

Image source: Ideogram / The Rundown

The Rundown: The Rundown Roundtable is a weekly feature where we poll members of The Rundown staff about how we use AI in our work or daily lives.

Jason, Developer: I used /goal in OpenAI’s Codex to build a Magic: The Gathering app so my brother and I can play asynchronously without needing to coordinate a call or awkwardly play over FaceTime.

The idea is to let each of us take turns when we have time, track the board state cleanly, and keep a game going over days instead of trying to line up schedules. The command allowed Codex to continue running until everything was done, basically one-shotting exactly what I was looking for without any intervention.

Joey, Partnerships: I've never been to Greece, so for my upcoming trip, I went all in and handed the whole itinerary over to Claude. Flights booked, transit times dialed, restaurant lists curated city by city.

I’m now showing up with a plan tighter than most travel agents could put together!

AI TRAINING

✅ Automate any manual task with Codex

The Rundown: In this guide, you will learn how to let Codex click through any annoying, repetitive work using Computer Use on Mac or Windows.

Step-by-step:

  1. Open Codex, go to Plugins, find and enable the Computer Use plugin, and start a new task

  2. Open the permissions menu and switch from Default permissions to Full access, then confirm any prompts and give Codex something real to do

  3. Example: “Open Chrome and debug this web page UI I’m developing http://localhost:3000/. Click through, reproduce the bug I describe, then tell me what you think is causing it. If not sure, ask before making changes”

Pro tip: Codex can automate repetitive workflows in local apps, too — try it for Photoshop exports, Adobe Premiere cleanup, file renaming, or any other tool.

PRESENTED BY ORACLE DEVELOPERS

🚀 Small models, bigger reasoning

The Rundown: Small language models can solve harder reasoning tasks without changing their weights. Oracle Developers’ open-source agent-reasoning code shows how to add research-backed orchestration to Ollama models, with 16 reasoning strategies developers can test locally.

In the guide, you’ll explore:

  • Open-source reasoning code for Ollama

  • 16 strategies, benchmarked across 4,200 runs

  • Better accuracy without retraining models

Get the open-source reasoning patterns. Explore the guide.

AI & ASTRONOMY

🪐 AI finds 100+ new exoplanets from NASA data

Image source: NASA

The Rundown: University of Warwick astronomers confirmed more than 100 exoplanets using an AI system called RAVEN that scanned 4 years of NASA TESS data covering 2.2M stars, with RAVEN also finding 2,000+ additional potential candidates.

The details:

  • RAVEN handles detection, vetting, and confirmation in one shot, trained on simulated planets and false-alarm signals to filter real finds.

  • The findings included 31 exoplanets never before spotted, plus strange worlds that orbit around their stars in under a day.

  • Hundreds of exoplanets were found in the "Neptunian Desert", a region so close to a star that Neptune-sized planets shouldn't survive the heat.

  • The system measures how common different planet types are at 10x the precision of previous systems from smarter AI alone, not new hardware.

Why it matters: Humans have confirmed just a few thousand exoplanets so far, and there are estimated to be trillions. AI and tech advances are going to help rewrite that number fast — and judging from RAVEN, all it will take is upgraded models and AI integrations to uncover knowledge about space already hiding in the data we have.

QUICK HITS

🛠️ Trending AI Tools

  • 🔒 Incogni - Remove your personal data from the web so scammers and identity thieves can’t access it. Use code RUNDOWN to get 55% off*

  • 💻 Codex in Chrome - OAI’s Codex extension for agentic tasks inside Chrome

  • 🧠 ERNIE 5.1 - Baidu's new foundation model with strong search capabilities

  • 🖨️ Printing Press - CLI factory with 30+ pre-built, agent-native tools

*Sponsored Listing

📰 Everything else in AI today

Google’s Isomorphic Labs is reportedly raising $2B+ to expand its Drug Design Engine, which it says significantly outperforms AlphaFold 3 on specific tasks.

Greece is proposing AI protections into its constitution, requiring the tech to serve individual freedom, with PM Mitsotakis citing threats to democracy.

Baidu released ERNIE 5.1, a new AI ranking No. 4 on Arena's Search Leaderboard, with the company claiming it cost just 6% as much to train as rival models.

OpenRouter launched Pareto Code, a free routing layer that auto-picks the cheapest coding AI above a user-set quality bar, with prices adjusting as newer models improve.

SoftBank Group’s telecom arm launched a battery business to build large-scale cells and storage systems — and meet the power demands of data centers in development.

COMMUNITY

🤝 Community AI workflows

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Anonymous:

"I have been using ChatGPT for various things professionally, which has been surprisingly helpful and refreshing. The greatest use I have found for it so far, though, is helping me train my 4 dogs.

I was ready to drop thousands of dollars on a professional trainer just because of how chaotic it has been, but ChatGPT has helped me identify the root causes of specific behaviors and taught me how to successfully train around and beyond them using specific techniques tailored to my individual dogs.

The confidence it has brought me, and the positive reinforcements have changed every dynamic in the household, and I wish I had started sooner."

How do you use AI? Tell us here.

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — the humans behind The Rundown

Tech

'RAMageddon' is coming for your laptop

Jennifer Mossalgue • 5 minutes

Read Online | Sign Up | Advertise

Good morning, tech enthusiasts. “RAMageddon” is the new name for an old tech nightmare: too much demand chasing too little silicon. As AI giants vacuum up the world’s memory chips, the squeeze is starting to show up everywhere else — from pricier laptops to vanishing budget devices.

The result? A new pecking order in computing, where whoever can afford the RAM gets to shape the future.

In today’s tech rundown:

  • The RAM panic may kill cheap laptops

  • Colossal to resurrect extinct African antelope

  • SpaceX eyes a massive, $119B chip factory

  • Bumble to kill the swipe in new AI redesign

  • Quick hits on other tech news

LATEST DEVELOPMENTS

HARDWARE

☠️ The RAM panic may kill the cheap laptop

Image source: Images 2.0 / The Rundown

The Rundown: AI’s appetite for memory has turned boring old RAM into tech’s newest choke point — dubbed “RAMageddon” — threatening to make cheap laptops and phones relics of a more innocent era, CNET reports.

The details:

  • The data-center boom is soaking up DRAM, HBM, and storage supply, pushing memory makers toward high-margin AI customers over consumer devices.

  • Analysts expect the shortage to push device prices higher, with Gartner forecasting PC prices up 17% and smartphone prices up 13% in 2026.

  • Budget hardware gets hit first: Gartner says the sub-$500 entry-level PC segment could disappear by 2028.

  • The squeeze is rippling through the industry, with reports of pricier hardware as manufacturers either raise prices, delay products, or kill low-margin models.

Why it matters: For years, cheap memory made cheap computing possible; now AI is rewriting that bargain by turning a commodity component into a scarce strategic asset. The winners are hyperscalers with the cash to lock up supply, while everyone else gets to find out what happens when the AI boom starts draining the rest of the tech stack.

TOGETHER WITH FIN

🎤 Hello, Operator: A product launch from Fin

The Rundown: Running an AI-first support operation requires a new set of capabilities. You need new ways to manage work, measure performance, and continuously improve the customer experience. Join leaders from Fin on May 14 as they reveal Operator, a new product that transforms how you deliver customer experiences.

The session will cover:

  • A first look at a groundbreaking new product and the vision behind it

  • A live discussion with execs from Fin

  • A chance to network with industry leaders and peers if you’re in San Francisco

Register now to join in San Francisco or virtually.

COLOSSAL BIOSCIENCE

🦌 Colossal to resurrect extinct African antelope

Image source: Colossal Bioscience

The Rundown: Colossal Bioscience, the Dallas biotech behind the engineered dire wolf, has a new target: the bluebuck, an antelope that vanished from Southern Africa roughly 200 years ago.

The details:

  • Scientists extracted DNA from a museum specimen in Sweden to reconstruct the bluebuck genome, then identified the genetic variants behind its key traits.

  • They are now editing roan antelope cells with CRISPR to approximate the bluebuck’s makeup, with roan surrogates set to carry the resulting embryo.

  • The bluebuck joins a portfolio that includes the woolly mammoth, Tasmanian tiger, dodo, giant moa, and the dire wolf pups unveiled in April 2025.

  • Colossal has raised $555M in funding as of September 2025, drawing celebrity investors including Peter Jackson, Paris Hilton, Tom Brady, and Tiger Woods.

Why it matters: Colossal has no public reintroduction site, and the IUCN’s top antelope specialist questions how much the project would be seen as a conservation priority. Critics also note that what emerges won’t technically be a bluebuck — a stem cell scientist called Colossal’s previous animals “synthetic proxies.”

SPACEX

🚀 SpaceX eyes a massive, $119B chip factory

Image source: Mario Tama / Getty Images

The Rundown: SpaceX filed plans for a $119B chip mega-factory in rural Texas — and if it gets built, it will be one of the largest private semiconductor investments in history, producing compute for everything from AI data centers to Optimus robots.

The details:

  • SpaceX is proposing a new “Terafab” mega-factory in Grimes County, Texas, to secure an in-house supply of advanced semiconductors and compute.

  • Filings indicate an initial investment of around $55B, with a long-term buildout that could reach up to $119B if all phases are completed.

  • The facility would produce advanced semiconductors and computing hardware entirely in-house, reducing Musk’s dependence on TSMC and Samsung.

  • Terafab is designed to supply the full Musk industrial stack: AI data centers, Starlink satellites, robotaxis, and Optimus humanoid.

Why it matters: Musk has been explicit about his reasoning: existing chipmakers aren’t expanding fast enough to meet his companies' AI and robotics needs. If Terafab gets built, it would be among the largest private semiconductor investments ever undertaken — though at this stage, it remains a proposal, not a sure thing.

BUMBLE

🐝 Bumble to kill the swipe in new AI redesign

Image source: Images 2.0 / The Rundown

The Rundown: Bumble’s CEO Whitney Wolfe Herd told Axios that the app is ditching the swipe-based interface that built its brand and betting that an AI-driven overhaul can reverse a brutal revenue slide.

The details:

  • The overhaul, dubbed “Bumble 2.0,” will lean on AI-driven recommendations to encourage more meaningful connections and actual dates.

  • In Q1 2026, Bumble’s paying users fell 21% year over year to about 3.2M, while overall revenue dropped roughly 14% to around $212M.

  • Herd is framing the decline as a strategic “reset” toward a smaller base of more engaged, higher-value members.

  • Match Group's Tinder and Hinge are both road-testing AI features and experimenting with new paid tiers to slow churn.

Why it matters: The dating app sector has spent years wondering whether the swipe behavior trained users to treat people like products. Bumble is now betting it did — and that tearing out its signature mechanic is the way forward. Whether Bumble 2.0 is a genuine reinvention or a rebranding of decline depends on what the AI actually delivers.

QUICK HITS

📰 Everything else in tech today

Apple is reportedly close to bringing camera‑equipped AirPods into production, turning its earbuds into a wearable that can see and interpret a user’s surroundings.

Volkswagen overtook Amazon as Rivian’s largest shareholder, boosting its stake to 15.9% through a multibillion-dollar joint venture focused on EV software.

Kalshi raised $1B in new funding at a $22B valuation, putting the prediction market platform roughly $10B ahead of DraftKings’ market value.

Kids are reportedly defeating age-verification systems on adult websites by simply drawing on fake mustaches with makeup pencils.

Utah passed a law that forces adult websites to verify the ages of anyone physically in the state and bans using or promoting VPNs to bypass those age checks.

Tin Can, the viral screenless phone for kids, now offers a “Communities” bulk-order program that lets groups buy discounted devices in large quantities.

Utah State University researchers showed that a CRISPR system called Cas12a2 can be programmed to precisely destroy cancer‑causing cells while sparing healthy cells.

COMMUNITY

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — The Rundown’s editorial team

AI

OpenAI closes reasoning gap in voice agents

Zach Mink • 7 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. Typing made AI useful, but speech is where agents have to prove they can keep up with real life.

OpenAI's new real-time voice model trio is built for that messier interface, adding a major reasoning upgrade, the ability to talk while thinking, and capable tool use that moves AI voice agents closer to running tasks at the speed of natural conversation.

In today’s AI rundown:

  • OpenAI’s reasoning upgrade for voice agents

  • Google folds Fitbit into its AI health play

  • Test multiple AI models with same prompt

  • Anthropic plans for AI that builds itself

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

OPENAI

🗣️ OpenAI’s reasoning upgrade for voice agents

Image source: OpenAI

The Rundown: OpenAI just introduced GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper, three API voice models that bring new reasoning, streaming, tool use, realism, and more capability upgrades to AI voice agents and live speech.

The details:

  • Realtime-2 brings GPT-5-level reasoning to live speech, is able to use multiple tools at once, talks while it thinks, and has better tone control for realism.

  • On Big Bench Audio, Realtime-2 hit 96.6% vs. 81.4% for its predecessor, a 15-point jump in how well voice AI can reason in real-time.

  • OpenAI also shipped a live translator covering 70+ languages and a streaming transcription model, rounding out a full voice-agent toolkit.

  • OAI said Zillow, Priceline, and Deutsche Telekom are already building on the models for real estate AI agents, voice-managed travel, and customer support.

Why it matters: AI voice’s turn-based era appears to be nearing a close, with OAI’s new model moving to systems that can reason better, leverage tools, and complete workflows without awkward interruptions that take users out of a natural flow. The AI industry is fixated on text agents, but the next wave will be spoken to, not typed at.

TOGETHER WITH AWS MARKETPLACE

📊 15+ enterprise leaders on getting data AI-ready

The Rundown: AWS Marketplace just released a free book featuring 15 chapters from senior data and AI leaders at JPMorgan Chase, Siemens, Mercedes-Benz, Roche, and more — each sharing practical advice on building the data infrastructure needed for agentic analytics and intelligent agents.

Chapters cover topics including:

  • Evolving data strategy for agentic AI and scaling data products

  • Building on existing infrastructure with a pragmatic, business-first approach

  • Unlocking value with classical ML, semantic layers, and cross-team alignment

  • Real-world perspectives from leaders across different industries

Get your free digital copy today.

GOOGLE

⌚️ Google folds Fitbit into its AI health play

Image source: Google

The Rundown: Google opened its AI health coach to the public after months in beta, integrating the Fitbit app into a new Google Health platform and pairing it with a new $99 screenless tracker that tracks and transmits body data to the AI.

The details:

  • Running on Gemini, the AI coach can tailor weekly workout routines, interpret uploaded medical records, and ID what a user ate from a phone photo.

  • Google is consolidating the Fitbit app, Health Connect, Apple Health, wearable data, and U.S. medical records into a single Google Health hub.

  • The new $99 Fitbit Air has no screen and weighs just 12g, carrying heart rate, oxygen, and temperature sensors that provide body data to the AI coach.

  • Apple Watch, Garmin, and Oura owners are set to get AI coach access later this year, with Google opening it up to hardware outside of its own.

Why it matters: AI’s role in personal health is only growing, and integrating everything under one roof can help Google make the AI layer the core product while also owning a trusted wearable line that provides users with the personalized guidance and context typically missing from other trackers and less connected options.

AI TRAINING

✏️ Test multiple AI models with same prompt

The Rundown: In this guide, you will learn how to use OpenRouter Fusion to test the same prompt across multiple AI models at once. Instead of opening five apps and guessing, you can compare outputs side by side and build a quick cheat sheet for work.

Step-by-step:

  1. Create an OpenRouter account, open OpenRouter Fusion, and pick how you want to pay for AI usage — OpenRouter credits or API keys you already pay for

  2. In Fusion, pick the models you want to compare — we tested Opus 4.7 vs. GPT 5.4 vs. Grok — and run one benchmark prompt at a time, keeping it identical

  3. Prompt something like: “You are advising a 20-person SaaS company deciding whether to replace its weekly status meeting with an async written update. Write a recommendation memo with 3 benefits, 3 risks, and a 2-week implementation plan. Keep it concise and practical”

  4. Open the responses, read the side-by-side analysis, and note which model is strongest. In the demo, about 10 comparisons cost around 40 cents

Pro tip: Run a few prompts you use all the time, write which model wins each task, and use OpenRouter's model browser to compare price and speed before you spend more.

PRESENTED BY WEIGHTS & BIASES

🐝 New guide: Tools and workflows to develop AI agents

The Rundown: AI agents can dramatically boost productivity and innovation, but getting them into the real world takes a lot of iteration. Whether you’re exploring agents for the first time or refining your current approach, this primer delivers actionable insights to help your team succeed and thrive in the AI era.

Get the guide to learn:

  • What defines agentic applications and why observability matters

  • A proven workflow for building agentic AI applications

  • How pioneering companies are building and deploying AI agents today

Download a primer on building successful AI agents.

THE ANTHROPIC INSTITUTE

🔬 Anthropic plans for AI that builds itself

Image source: Anthropic

The Rundown: Anthropic's newly formed research arm, The Anthropic Institute, published its formal research agenda — a document that treats the possibility of AI systems improving themselves as something the company is actively preparing for.

The details:

  • TAI sits inside Anthropic, letting researchers study Claude usage, internal workflows, and security signals before they hit the wider market.

  • The Institute’s agenda spans security threats, economic disruption, governance, and planning for self-improving models.

  • The team also proposed Cold War-style hotlines between labs and governments, plus "fire drill" exercises for sudden capability surges.

  • TAI said it is committed to publishing Economic Index data, monthly worker surveys, threat research, and more details on its own internal AI-boosted R&D.

Why it matters: We wrote earlier about Anthropic co-founder Jack Clark’s blog on self-improving systems, and TAI’s research agenda puts it very much into focus. Anthropic’s talk of “fire drills” and Cold War-style systems is to prepare for an “intelligence explosion” that we might be heading to faster than many expected.

QUICK HITS

🛠️ Trending AI Tools

  • ✈️ Serko.ai - The AI travel assistant that plans, books, and manages your entire business trip, so you can skip the busywork*

  • 🗣️ GPT-Realtime-2 - Voice AI that thinks, calls tools, maintains convo flow

  • 🎥 Studio Agent - ElevenLabs' AI editor to draft videos, places sound effects

  • 🎆 Grok Imagine Quality Mode - xAI’s Image generation with higher realism

*Sponsored Listing

📰 Everything else in AI today

Spotify launched ‘Personal Podcasts’, a tool allowing agents to turn items like briefings or class notes into a personal podcast directly inside users’ Spotify libraries.

OpenAI introduced Trusted Contact, an opt-in ChatGPT feature that alerts a designated friend or family member if signs of self-harm risk are detected.

Scale AI landed a $500M Pentagon contract for military data analysis, marking a 5x jump from last September's $100M deal.

Perplexity rolled out its Personal Computer to all Mac users, allowing it to take agentic action across a user’s local computer, files, and via the Comet browser.

Mozilla published a blog about using Claude Mythos Preview for security, saying the model patched more bugs in April than the past 15 months combined.

COMMUNITY

🤝 Community AI workflows

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Tatiana B. in San Francisco, CA:

"I'm COO of a startup and mom to a two-and-a-half-year-old. Managing both is a lot, and keeping track of everything I need to do at home on top of work can be a real mental drain.

So I use AI to help me compile a document covering everything I need help with at home: my daughter's meal preferences, her daily routine, and house chores. I treat it like a work project, going back and forth with AI to think through what I actually need done, fill in gaps I hadn't thought of, and get it all out of my head into something I can hand to someone else.

Now, when someone new comes to help at home, I don't have to explain everything from scratch. That frees me up to actually be present with my daughter when I'm with her, and focused on work when I'm not. I also know I'm lucky to be in a position to hire help, but using AI to think clearly about what you need and get it out of your head is something anybody can do."

How do you use AI? Tell us here.

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — the humans behind The Rundown

No matching search results

Try using different keywords, double-check your spelling, or explore related categories.

Clear Search

Stay Ahead on AI.

Join 2,000,000+ readers getting bite-size AI news updates straight to their inbox every morning with The Rundown AI newsletter. It's 100% free.

Subscribe