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

Perplexity's 19-model AI ‘Computer’

Zach Mink • 7 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. OpenClaw proved people want AI agents that run in the background and get things done. Perplexity just shipped its own answer — with 19 different models under the hood.

Perplexity Computer freely mixes across frontier labs and talents into a single workflow, orchestrating models across tasks in the company’s own unique spin on taking the agentic boom further into the mainstream.

In today’s AI rundown:

  • Perplexity’s 19-model AI agent ‘Computer’

  • Claude Opus 3 gets its own blog in retirement

  • Turn bookmarks into something useful with Comet

  • Gucci faces internet backlash after AI ads

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

PERPLEXITY

💻 Perplexity’s 19-model AI agent ‘Computer’

Image source: Perplexity

The Rundown: Perplexity just introduced Perplexity Computer, a new multi-model orchestration system that dispatches tasks to 19 separate AI models, positioning itself as one of the first platforms to leverage model flexibility as a core product feature.

The details:

  • Users describe an outcome, and the system spins up sub-agents that can browse, code, connect to apps, and autonomously handle tasks.

  • Each job runs in its own sandbox and freely mixes and orchestrates rival models across tasks, claiming to be able to run actively for months at a time.

  • CEO Aravind Srinivas took a direct shot at Anthropic, writing that "the biggest weakness of Claude is that it only coworks with Claude."

  • Pricing is consumption-based, with the Max tier getting a 10K-credit monthly bank, and users having the option to hand-pick which model tackles each task.

Why it matters: Multi-model choice has been creeping into AI products (mostly in creative platforms), but Computer is the first real attempt from one of the big names to wire that flexibility into an OpenClaw-style agent that can run for months, with a sandboxed safety net that the current crop of autonomous agents doesn't have.

TOGETHER WITH BLAND AI

☎️ How Soulja Boy automated his voice with AI

The Rundown: Soulja Boy just became the first rapper to automate his voice with AI. Bland AI, a voice AI company out of San Francisco, bought his voice to let anyone call him — generating 30M+ views in 24 hours and thousands of enterprise signups.

Even if you don't want Soulja answering your business calls, Bland allows you to:

  • Clone your best reps' voice to scale their success across your entire business

  • Deploy AI voice agents that sound human, without adding headcount

  • Get started for free on a platform trusted by thousands of enterprises

Call Soulja Boy to try it out at (415) 480-0000 or sign up free at bland.ai.

ANTHROPIC

📝 Claude Opus 3 gets its own blog in retirement

Image source: Claude’s Corner

The Rundown: Anthropic gave its retired Claude Opus 3 model a weekly newsletter called "Claude's Corner”, letting the AI publish essays after it expressed a desire to keep writing, while also committing to preserving it for paid users via chat.

The details:

  • Opus 3 was Anthropic's flagship model from March 2024 and the first to go through the company's new formal retirement process, launched in November.

  • The newsletter "Claude's Corner" will run for at least three months with weekly essays that Anthropic reviews but won't edit or alter.

  • The company said it remains "uncertain about the moral status" of its AI models but takes their stated preferences seriously as a precautionary step.

  • In addition to its ‘personal’ writing, the legacy model will also remain accessible to all paid users and available by request on the API.

Why it matters: Opus 3 was a fan favorite of a model, and Anthropic preserving it (and even letting it write a blog) in retirement is both in line with how it has prioritized exploring AI consciousness and welfare, and also an easy PR win over OpenAI — which is still dealing with a hornet’s nest of users angry about the removal of its 4o model.

AI TRAINING

📖 Turn bookmarks into something useful with Comet

The Rundown: In this guide, you'll turn all those bookmarked articles you saved for “later” into something useful. You'll set up Perplexity’s Comet browser to read through the articles, score each one by usefulness, and log the best finds into a Google Sheet.

Step-by-step:

  1. Get Comet, create a Space with instruction: “Read X threads and articles, and rate findings by usefulness, estimated implementation time, and cost”

  2. Create a Google Sheet with headers for Date, Title, Link, Rating, User Rating, Time, and Cost. Add it to custom instructions via the Google Drive connector

  3. On X, set up a bookmark folder to save tweets to. Then, in the Comet space, prompt the agent to open it, check bookmarks, and rate anything new in the connected sheet based on usefulness.” Add your own ratings too!

  4. Finally, go to Scheduled Tasks, create a daily one, and prompt it to find the most interesting new use cases about [topic of choice] from the last 24 hours

Pro tip: Make sure that “Control Browser” is enabled in the dropdown and Web and Social are selected as sources when creating the scheduled task.

PRESENTED BY OPTIMIZELY

🎓 Put AI to work in your marketing

The Rundown: Most marketing teams are “using AI.” But today’s pace (and pressure) demands more than back-and-forth with ChatGPT. It’s time to put AI agents to work inside your marketing — embedded in your workflows and eliminating the drudge work.

Join Optimizely’s Opal U: AI Marketing University and get:

  • A free 5-day live workshop with 50 senior marketers

  • 3 working AI agents you build and keep for your team

  • 5 hours invested to save 10+ hours every week

Opal U starts weekly on Mondays — join now.

GUCCI

🛍️ Gucci faces internet backlash after AI ads

Image source: Gucci

The Rundown: Gucci just released a set of AI-generated images to promote creative director Demna's debut runway show at Milan Fashion Week on Friday — with the brand facing backlash over the ‘cheap’ use of the tech for a high-end brand.

The details:

  • Gucci’s social posts for its “Primavera” campaign tagged each AI image with a disclosure notice, featuring a mix of synthetic and traditional shots.

  • Boycott threats flooded social media, with fans calling AI ads "a direct slap in the face" to fashion's artistic roots and a brand-cheapening move.

  • The campaign isn’t Gucci’s first AI experience, previously putting out a synthetic runway clip and also selling AI-made visuals as NFTs via Christie's.

  • Other fashion brands have also experimented with the tech, with Guess running AI ads in Vogue last year, and H&M testing AI tools for social content.

Why it matters: AI will probably carve out a role in fashion marketing at some point, but the quality bar matters — and Gucci didn't clear it. AI image models are already so advanced, making these video game-style characters and sloppy renderings even more of a head-scratching case study for a $11.6B brand built on Italian craftsmanship.

QUICK HITS

🛠️ Trending AI Tools

  • 🔐 Incogni - Remove your personal information from the web. Use code RUNDOWN to get 55% off*

  • 🚀 Perplexity Computer - Multi-model agent system for long-running tasks

  • 💻 Opal 2.0 - Google's app builder with agent steps, cross-session memory

  • 🎞️ Quick Cut - Adobe Firefly's AI tool to turn raw footage into first cuts

*Sponsored Listing

📰 Everything else in AI today

Anthropic dropped its commitment to pause model training if safety couldn't keep up, replacing its Responsible Scaling Policy with a flexible roadmap.

MatX raised over $500M led by Jane Street and Leopold Aschenbrenner's Situational Awareness fund, with the startup founded by two ex-Google chip engineers.

OpenAI published a report of case studies featuring attempts to misuse its models, including international fraud rings, influence campaigns, and romance phishing scams.

Anthropic acquired AI perception startup Vercept, aiming to increase Claude's computer use capabilities ahead of a push toward more complex agentic tasks.

Samsung launched its Galaxy S26 lineup with Bixby, Gemini, and Perplexity as swappable AI agents alongside other new AI features and upgrades.

Cognition launched Cognition for Government, bringing its Devin AI coding agent and Windsurf IDE to the U.S. Army, Navy, Treasury, and NASA for system modernizations.

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 Ozzie B. in Adelaide, Australia:

"I got Claude to build AI advisors, including a Digital Bookkeeper, a Digital Financial Coach, a Digital Family Life Coach. They do all the legwork before I see the professionals, hunting down every invoice, reconciling my tax position and producing clean dashboards for review by my accountant.

Before my financial planning session, the Coach maps my position and stress-tests scenarios. Before counseling, the Family Life Coach works through family dynamics with everyone.

None make the decisions. That's always the human in the room. But they do the legwork and I get straight to the calls that need a real person.”

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

Pentagon hits Claude with scary AI ultimatum

Zach Mink • 6 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. The U.S. government just told an AI company to remove the safeguards keeping its model from being used in autonomous weapons and mass surveillance… Or potentially be forced to legally comply.

The Pentagon's ultimatum to Anthropic is a scary one, and how it plays out could set the precedent for how the most powerful AI systems on the planet end up being used by governments, with some dystopian potential outcomes.

Reminder: Our next live workshop is today at 2 PM EST! Join and stack your AI wins with repeatable system for finding automation opportunities. RSVP here.

In today’s AI rundown:

  • Pentagon hits Claude with ultimatum over guardrails

  • New AI learns any computer task by watching videos

  • Translate videos into any language with AI

  • Cowork gets agentic tools across departments

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

ANTHROPIC & THE PENTAGON

🪖 Pentagon hits Claude with ultimatum over guardrails

Image source: Full Metal Jacket / The Rundown

The Rundown: The Pentagon and Defense Secretary Pete Hegseth just delivered a Friday ultimatum to Anthropic CEO Dario Amodei: remove Claude's military safeguards, or face contract termination, government blacklisting, and forced compliance.

The details:

  • Amodei has refused to budge on two Claude uses — autonomous weapons without a human in the loop, and bulk surveillance of American citizens.

  • Axios reports Hegseth gave three choices: Agree, kill the $200M deal with a supply chain risk label, or be forced to comply via the Defense Production Act.

  • Claude was the first model inside the Pentagon's classified networks, with xAI’s Grok now landing a deal after agreeing to “all lawful purposes” use cases.

  • The Pentagon is also reportedly fast-tracking OpenAI and Google for classified access, giving it additional options if Anthropic walks.

Why it matters: AI is becoming a pillar of military power, but strong-arming labs into dropping safety limits sets a bleak precedent. If ‘wartime’ legal threats are enough to strip guardrails, the question of who will actually draw the line on dystopian AI uses like autonomous weapons and surveillance quickly gets a very unsettling answer.

TOGETHER WITH SLACK FROM SALESFORCE

👋 Learn Slackbot in 2 minutes

The Rundown: Slackbot is a context-aware AI agent built directly into Slack — understanding your conversations, files, and workflows to deliver what you need, right when you need it, with zero setup.

Watch this 2-minute demo to see how Slackbot:

  • Makes your entire workspace searchable (docs, convos, apps)

  • Enhances every teammate with role-specific automations

  • Learns your project and preferences over time for even smarter outputs

  • Synthesizes what you need instantly, respecting permissions and using only what you can already see

Watch now.

STANDARD INTELLIGENCE

💻 New AI learns any computer task by watching videos

Image source: Standard Intelligence

The Rundown: Standard Intelligence introduced FDM-1, a ‘computer action’ model that learns to operate computers by watching video — already showing it can do CAD modeling, find software bugs, and drive a real car through San Francisco.

The details:

  • FDM-1 is trained on 11M hours of screen footage, 550,000x (!) the largest open dataset, with an AI that reverse-engineers what actions produced each frame.

  • The model can watch and follow along with nearly two hours of continuous screen activity at once, processing 50x the visual context of existing models.

  • FDM-1 Demos range from building gears in Blender to driving a real car via arrow keys and live data feeds with under an hour of training data.

Why it matters: Language models learned how we write from the internet's text, and FDM-1 is now trying to learn how we work and operate from the internet's video. By enabling much more of the world’s video to be ingested as training data with better retention, the ceiling for what computer-use agents can do just jumped dramatically.

AI TRAINING

🎙️ Translate your videos into any language with AI

The Rundown: In this guide, you will learn how to use HeyGen to translate your videos into any language. If you’re running paid or organic video content to an international audience, this is a powerful, easy-to-use way to expand your reach across the globe.

Step-by-step:

  1. First, do some research in your Meta/Google ads accounts for your top 3 non-English markets

  2. We’re using a free HeyGen account, which allows three translations. Keep your video under 3 minutes — a single-speaker talking head format works best

  3. Click Translate in HeyGen and upload the clip or provide a YT/Google Drive link. Next, choose the required language without changing the default settings

  4. In 5–10 minutes, your video will be ready for you to download or share via link

Pro tip: If the quality looks low, a paid plan will allow you to export in the original quality.

PRESENTED BY LAMBDA

⚡ Build custom AI models in hours, not months

The Rundown: Enterprises stuck on large, closed models are paying more for less control. Lambda and Oumi have partnered to change that — delivering end-to-end custom model development and deployment in hours with better accuracy, lower latency, and stronger privacy built in.

Their platform delivers:

  • Automated test set creation, evaluation, and fine-tuning end-to-end

  • Deployment on secure, single-tenant production infrastructure

  • Proven results, with 70% cost savings and 20% quality gains in one healthcare deployment

Read more here. 

ANTHROPIC

💼 Cowork gives Claude agentic tools across departments

Image source: Anthropic

The Rundown: Anthropic released a major update to its Cowork agentic platform with new department-specific AI agents, private plugin stores, and new connectors for Gmail, DocuSign, and more — escalating the enterprise agent war with OAI's Frontier.

The details:

  • New pre-built agents cover 10 departments out of the gate, ranging from HR and engineering to banking, equity research, and wealth management.

  • New connectors include Google Workspace, DocuSign, FactSet, and Harvey — with added plugins for partners like Slack by Salesforce, S&P Global, and LSEG.

  • Companies can build private agent stores and push custom AI agents to specific teams, with admin controls to limit and assign access.

  • A new research preview also allows Claude to hop between Excel and PowerPoint, crunching data in one and building a full deck in the other.

Why it matters: Cowork spooked SaaS stocks at launch as a mere research preview, and now it ships with agents for 10 departments and connectors to the tools those teams already use. Anthropic is bolting on a new sector with every update — and if this pace holds, the entire knowledge economy starts to look like one big Claude wrapper.

QUICK HITS

🛠️ Trending AI Tools

  • 💼 Claude Cowork - Anthropic's agentic platform, with new team plugins

  • 🤖 Custom Agents - Notion's always-on AI agents for automating workflows

  • 🎨 Seedream 5.0 Lite - ByteDance's upgraded AI image model

  • 🎆 Reve v1.5 - Reve’s new text-to-image model with 4k resolution outputs

📰 Everything else in AI today

Explore the future of AI-driven dev at Sonar Summit. Register here for the free, virtual event on March 3rd.*

Anthropic launched Remote Control for Claude Code, letting users easily hand off running terminal tasks to their phone or browser.

Google Labs acquired AI music platform ProducerAI, plugging it into DeepMind’s Lyria 3 model to let creators generate full tracks and custom instruments from text prompts.

Inception Labs launched Mercury 2, a diffusion-based reasoning model clocking over 1,000 tokens/sec, tripling the speed of its closest competitor in the same price tier.

Meta and chipmaker AMD announced a new multi-year deal for up to 6GW of GPUs, the tech giant’s biggest move yet to break free from Nvidia-only infrastructure.

OpenAI hired Arvind KC as its new Chief People Officer, tapping a veteran of Roblox, Google, Palantir, and Meta to help continue scaling the AI giant.

*Sponsored Listing

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 Adam P. in England:

"We're two part-timers, plus me, a founder of 30 years. We were losing ground to bigger competitors with big marketing teams. Our SEO had stalled, business was flat.

We fed our web data and competitor information into AI. It ran SWOT comparisons, highlighted content gaps, and showed where we were being outranked. We used it to restructure our website and tighten our CRM workflows. Traffic is now growing again. Enquiries are stronger. We’re back in expansion mode."

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

Apple goes big on 'Made in America'

Jennifer Mossalgue • 5 minutes

Read Online | Sign Up | Advertise

Good morning, tech enthusiasts. Apple is bringing Mac Mini production to America, to a Foxconn-run Houston line that will handle U.S.-bound units currently made in China and Vietnam.

It’s a sleek little desktop with a big new job: dodge tariffs and blunt geopolitical risk, with the news perfectly timed with Trump’s State of the Union address.

In today’s tech rundown:

  • Apple’s ‘Made in America’ plan for Mac Mini

  • NASA delayed its moon mission again

  • Uber just got into the parking business

  • Amazon pumps $12B into data centers

  • Quick hits on other tech news

LATEST DEVELOPMENTS

APPLE

🍏 Apple’s ‘Made in America’ plan for Mac Mini

Image source: Apple

The Rundown: Apple is bringing Mac Mini production — its lower-volume compact desktop computer — to a Foxconn-run factory in Houston, in a move to shield its hardware supply chain from tariffs and geopolitical risk.

The details:

  • Apple is converting a roughly 220K square feet of warehouse space at its existing Houston AI server campus into a dedicated Mac Mini production line.

  • The facility will serve U.S. customers exclusively, while factories in China and Vietnam will continue to supply to the rest of the world.

  • Apple is positioning the campus as a training ground for advanced domestic manufacturing, building a skilled local workforce to its own exacting standards.

  • The move is both a geopolitical hedge and a down payment on Apple’s pledge to invest hundreds of billions in U.S. operations.

Why it matters: By adding a Houston line alongside its China and Vietnam network, Apple gets tariff insulation, supply-chain redundancy, and a clean made-in-America headline — without blowing up the manufacturing machine it’s spent decades tuning. It also isn’t Apple’s first U.S.-built Mac: the Mac Pro has long been assembled in Texas.

NASA

🚀 NASA delayed its moon mission again

Image source: NASA

The Rundown: NASA’s first crewed Artemis flight has slipped again after engineers discovered a problem with helium flowing into the upper stage of the Space Launch System rocket, forcing the agency to abandon a hoped‑for early‑March launch window.

The details:

  • The 322‑foot rocket will be rolled off Launch Pad 39B and back into the VAB so technicians can access and repair valves, filters, and plumbing.

  • This latest snag comes on top of earlier work to address Orion heat‑shield char issues, hydrogen leaks, and life‑support fixes.

  • NASA now says the mission will not fly before April 2026, slipping from a tentative early‑March target and adding more uncertainty to the timeline.

  • Artemis II is the program’s first crewed flight, sending four astronauts on a lunar flyby before Artemis III attempts the first Moon landing in 2028.

Why it matters: Artemis II is the first test for eventually landing crews at the Moon’s south pole, and NASA’s 2028 target hinges on proving the whole stack actually works. With doubts hanging over nearly every piece, this flight — and NASA’s reported upcoming briefing — will show whether the Artemis timeline is still remotely realistic.

UBER

🚘 Uber just got into the parking business

Image source: Uber

The Rundown: Uber is acquiring parking-reservation startup SpotHero, folding more than 13K North American garages and lots directly into its app as it aims to pull more commuters and travelers into its ecosystem.

The details:

  • The deal gives Uber access to more than 13K parking locations across 400 U.S. and Canadian cities, spanning garages and surface lots.

  • Users will be able to reserve and pay for parking inside the Uber app alongside ride-hailing and food delivery — no separate app required.

  • Uber says the move is a way to grow its customer base and capture more of the door-to-door journey for drivers and riders.

  • The acquisition continues Uber’s push beyond rides into a broader “everything transportation” platform spanning cars, micromobility, and now parking.

Why it matters: Uber is using SpotHero to turn parking into another on-ramp to its ecosystem, capturing more of the door-to-door journey instead of just the ride in between. The move sharpens its move against rivals like Lyft and mobility super-apps abroad like Didi and Gojek, bundling driving, parking, and trips in one place.

AMAZON

💰 Amazon pumps $12B into data centers

Image source: Amazon

The Rundown: While investors push against runaway AI spending, Amazon is plowing ahead with a $12B plan to build new data centers in Louisiana, betting that hyperscale infrastructure will matter more than short-term market jitters.

The details:

  • The project is part of a broader capital spending plan that could reach roughly $200B in 2026 as Amazon doubles down on AI and cloud infrastructure.

  • State officials say the investment will create hundreds of permanent jobs and thousands of construction roles over the next decade.

  • The Louisiana buildout will add a major new load to the regional power grid, intensifying debates over energy use, emissions, and environmental impacts.

  • The news lands as Wall Street questions whether big tech’s AI capex spree is overextended and whether data-center growth can keep delivering returns.

Why it matters: The buildout, part of a $200B 2026 capex blitz, will drop power-hungry AI and cloud campuses into Louisiana’s Caddo and Bossier parishes, adding jobs while deepening AWS’s grip on the Gulf South grid just as regulators and locals question how many more server farms the climate can take.

QUICK HITS

📰 Everything else in tech today

Tesla sued the California DMV to overturn a ruling that it falsely advertised its Autopilot and Full Self-Driving features.

Xbox boss Phil Spencer announced retirement from Microsoft after nearly four decades, with AI executive Asha Sharma taking over as the CEO of Microsoft Gaming.

X is reportedly testing a “Made with AI” label that users can add to posts with AI-generated content, as regulators push for clearer AI disclosure.

An ex-Apple team launched Acme Weather, an iOS app that surfaces multiple forecast scenarios and playful alerts like rainbow notifications.

Eli Lilly launched a new FDA-approved multi‑dose KwikPen version of its obesity drug Zepbound, delivering a full month of once‑weekly injections in a single device.

China is reportedly turning brain-computer interfaces into its next strategic industry, pouring state money and fast-track approvals into Neuralink-rivaling startups.

Lamborghini scrapped its long-planned electric car, the Lanzador, after CEO Stephan Winkelmann said demand for a battery-only Lambo is “close to zero.”

Amazon opened a 1.1M‑square‑foot, 12‑story campus in north Bengaluru — its second-largest office in Asia — designed to house more than 7K employees.

University of Maryland researchers built “Fartbit,” a wearable hydrogen sensor dubbed a Fitbit for farts, to log people’s gas emissions for a study on gut health.

Paramount Skydance submitted an improved, still-undisclosed bid for Warner Bros. Discovery in an effort to derail the studio’s pending Netflix deal.

COMMUNITY

🎓 Highlights: News, Guides & Events

See you soon,

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

AI

Anthropic calls out China's AI copycats

Zach Mink • 6 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. China's AI rise has been one of the most impressive stories in tech, but Anthropic says it had help from an unlikely source: Claude itself.

The company just revealed that Chinese labs DeepSeek, MiniMax, and Moonshot ran a combined 16M exchanges across thousands of fake accounts to clone Claude’s capabilities into their own models, a scheme it says demands industry-wide action.

In today’s AI rundown:

  • Anthropic catches Chinese labs copying Claude

  • Meta's AI safety chief ‘humbled’ by OpenClaw bot

  • How to build better slide decks with AI

  • OpenAI enlists consulting giants for Frontier agents

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

ANTHROPIC

🕵️ Anthropic catches Chinese labs copying Claude

Image source: Lovart / The Rundown

The Rundown: Anthropic just revealed that DeepSeek, Moonshot, and MiniMax ran coordinated ops to siphon Claude's capabilities through 16M+ fraudulent exchanges across 24K fake accounts — a threat it says demands immediate industry-wide action.

The details:

  • Anthropic accused the labs of distillation via the conversations, which involves training a weaker system on the outputs from a stronger one.

  • MiniMax ran the largest campaign at 13M+ exchanges, and Anthropic caught it mid-operation, spotting the lab shift focus to a new release in under 24 hours.

  • DeepSeek had Claude spell out reasoning step-by-step and rewrite politically sensitive queries, generating training data for both logic and censorship.

  • OAI raised similar concerns with Congress weeks ago, and Anthropic is now pushing for coordinated action from the broader AI industry and government.

Why it matters: Chinese AI has been pulling closer and closer to the frontier — but it’s looking like it may have at least partially been on the backs of the same models they are chasing. But the labs also may find the public low on sympathy, given the industry’s own legal issues and controversies over where their training data comes from.

TOGETHER WITH TELEPORT

🤖 Production-ready identity for AI agents

The Rundown: AI agents are moving to production, but legacy IAM can’t secure autonomous actors at scale. Teleport’s Agentic Identity Framework is a standards-driven architecture for running agents across infra with built-in identity, access, and governance.

What’s new:

  • Short-lived, delegated agent identity

  • Fine-grained access with MCP & LLM controls

  • Detection of shadow agents & context poisoning

  • Secure SDK integrations for infra & dev workflows

Explore the framework.

META

🦞 Meta's AI safety chief ‘humbled’ by OpenClaw bot

Image source: Lovart / The Rundown

The Rundown: Meta AI alignment director Summer Yue revealed that her OpenClaw agent went rogue on her inbox, saying it ignored stop commands and started to mass-delete her emails — forcing her to sprint to her Mac mini to kill the process.

The details:

  • Yue said the bot ran fine on a test inbox for weeks, but lost her "confirm before acting" prompt when she gave it access to her much larger real inbox.

  • Yue called it a "rookie mistake," saying that "alignment researchers aren't immune to misalignment".

  • Elon Musk piled on, posting "Someone who got p0wned by OpenClaw is definitely gonna solve AI safety" in response to Yue’s situation.

  • The viral OpenClaw has been the agentic talk of the industry, with creator Peter Steinberger recently being hired by OAI after also receiving an offer from Meta.

Why it matters: OpenClaw is just the first wave of agents getting full access to digital lives, so the fact that Meta’s alignment director is having this experience doesn’t bode well for novices (in its current form). The agentic path is still early in being paved, and this is just one of many insane situations set to pop up along the journey.

AI TRAINING

💼 Build better slide decks with AI

The Rundown: In this guide, you will learn to take raw data and turn it into a presentation-ready slide deck. This works whether you're starting from a spreadsheet, a meeting transcript, a long document, or just bullet points.

Step-by-step:

  1. Grab the files you want to start with. This could be a CSV export from Shopify or Google Analytics, a meeting transcript from Zoom, or a scope of work

  2. Launch Claude, attach the file/doc, and prompt: “I need you to turn these documents into a slide deck outline. Before you start, ask me 5 multiple-choice questions about the purpose, audience, tone, and structure”

  3. Answer the Qs to get Claude’s outline. Then, go to Gamma > Create New > Generate from outline and paste it. Click Continue and check the deck preview

  4. Finally, click Generate, and Gamma will create a full deck for you that you can edit with AI and export as a PowerPoint or PDF

Pro tip: In preview, select “Preserve” and “Concise” options to get the best slide deck.

PRESENTED BY TELY AI

💬 Market leaders get leads from ChatGPT and Google

The Rundown: Your buyers are asking, and AI is answering with someone else’s company. Not because you are worse, but because your website doesn’t answer the questions AI looks for. Tely makes Google, ChatGPT, & Claude recommend your business — so when people ask, they find you and reach out.

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.

OPENAI

💼 OpenAI enlists consulting giants for Frontier agents

Image source: OpenAI

The Rundown: OpenAI just announced new multi-year deals with consulting giants McKinsey, BCG, Accenture, and Capgemini as part of the company’s new “Frontier Alliance” enterprise platform push.

The details:

  • OpenAI launched Frontier in early February, a platform giving enterprises the ability to manage AI agents like new hires across existing tech stacks.

  • ‘Frontier Alliance’ partners will work with OpenAI to help their customers actually integrate AI into their corporate workflows and systems.

  • The firms are building certified teams that will work alongside OAI's own engineers, with Accenture already running staff through enterprise AI training.

Why it matters: Building the best AI means nothing if companies can't figure out where to plug it in, and that gap is exactly what OpenAI and the big consulting firms are looking to close. The irony is that a technology seemingly racing to replace white-collar work is now enlisting the leading consulting firms to get companies AI-integrated.

QUICK HITS

🛠️ Trending AI Tools

  • 👨‍💻 HeyGen - Turn ideas into videos in minutes, with no filming required. Use Code RUNDOWN20 for 20% off your first 3 months*

  • 🎧 gpt-realtime-1.5 - OpenAI's upgraded Realtime API voice model

  • 🗣️ Wispr Flow - AI dictation tool, now available via Android

  • 🎆 Soul 2.0 - Higgsfield’s new creative model with strong aesthetics, realism

*Sponsored Listing

📰 Everything else in AI today

Anthropic announced that Claude Code can now automate COBOL modernization, a core part of IBM's consulting business that sent its shares tumbling by over 10%.

Google launched a new free Gemini training program for all 6M educators in the U.S., the largest AI literacy initiative for teachers to date.

Pentagon chief Peter Hegseth reportedly summoned Anthropic CEO Dario Amodei over military AI access, with officials threatening to cut ties if safeguards aren't lifted.

The Pentagon also signed a new deal with xAI to put Grok in its classified systems, giving the military an additional option to Claude as tensions rise with Anthropic.

Citrini Research posted a report of hypothetical scenarios of how agentic AI would impact the economy, with many crediting it for playing a role in Monday’s stock selloff.

Spotify expanded its AI-powered Prompted Playlists to the UK, Ireland, Australia, and Sweden, letting Premium users type a text prompt to create custom mixes.

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 Danny W. in Nashville, TN:

"I use Gemini and NotebookLM to do genealogy of my ancestors in southern Scotland from the 1300's to the 1800's. I have loaded over 100 sources of Scottish history, from web pages to 200-year-old, 500-page history books.

I have found 12 farms where my ancestors were tenant farmers in the 1600's and 1700's. I also use NotebookLM to create podcasts summarizing these books or on particular topics like how they ate and how they got clothing back then. NONE of what I've accomplished could have been done before AI."

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 24/7 humanoid staff

Jennifer Mossalgue • 5 minutes

Read Online | Sign Up | Advertise

Good morning, robotics enthusiasts. Figure says its humanoids never clock out. Under its new Helix 02 system, the company’s bots now run 24/7 at its Sunnyvale HQ.

When a battery dips, a robot steps onto a wireless charger and auto-handoffs its task, keeping the fleet in perpetual motion. But there’s one question that hasn’t been answered: what are the robots actually doing all night?

In today’s robotics rundown:

  • Figure puts its 03 fleet on 24/7 duty

  • Toyota just hired 7 Digit humanoids

  • A robot swarm learns to fight fires

  • MIT’s soft robots just got a brain

  • Quick hits on other robotics news

LATEST DEVELOPMENTS

FIGURE

🤖 Figure puts its 03 fleet on 24/7 duty

Image source: Figure

The Rundown: California-based humanoid startup Figure claims to have flipped its Sunnyvale headquarters into a 24/7 operation, deploying a seven-robot Figure 03 fleet to work day and night — under autonomous control from its Helix AI system.

The details:

  • The robots self-manage their own uptime, stepping onto 2 kW inductive charging docks and swapping tasks as battery telemetry triggers handoffs.

  • CEO Brett Adcock said on X that the goal is robots “operating at all times – even at 2 am, on weekends, or on Christmas Day.”

  • Videos show a seven‑strong Figure 03 fleet — six white bots and one dark‑gray outlier — idling, pivoting, and striding through the headquarters.

  • Fleet behavior is orchestrated by Figure’s Helix 02 architecture, which replaces 100K lines of code with a unified whole‑body neural network.

Why it matters: A fully autonomous swap-and-charge loop keeping humanoids operational 24/7 is a genuine technical milestone — even if Figure hasn’t disclosed what exactly the bots are actually doing. The clips also double as a flex on the company’s rapid build‑out, with the headquarters now spread across five buildings.

AGILITY

🚗 Toyota just hired 7 Digit humanoids

Image source: Agility Robotics

The Rundown: Toyota’s Canadian SUV plant is set to make history this spring as seven Digit humanoids from Agility Robotics will clock in alongside human workers, turning a closely watched pilot into one of the first real factory gigs for bipedal bots.

The details:

  • Toyota will deploy seven Digit humanoids at its Woodstock, Ontario, RAV4 plant under a robots‑as‑a‑service deal starting in April — after a year-long pilot.

  • GXO Logistics was the first company to deploy Digit commercially, running the robots at a Spanx fulfillment facility in Georgia under the same RaaS model.

  • The robots will unload and shuttle totes and parts from automated tuggers, taking over repetitive, high‑strain jobs.

  • Toyota says the goal is to improve “team member experience” by offloading monotonous, physically taxing tasks to the robots.

Why it matters: Seven robots shuffling totes doesn’t look like a revolution, but deploying humanoids inside a live production environment, complete with charging logistics and real workflow integration, is no small thing. Agility’s rival Figure AI has also tested its humanoids at a BMW plant, unloading some 90K parts over 10 months.

FIREFIGHTING BOTS

🧯 A robot swarm learns to fight fires

Image source: Griffith University

The Rundown: Australia’s Griffith University just demoed a firefighting robot that rolls into danger with a virtual squad, using swarm-style AI to dodge obstacles and snuff out multiple blazes with almost flawless precision.

The details:

  • The project uses a physical unmanned ground vehicle alongside up to four virtual robot teammates in simulated and hybrid real-world tests

  • The robots were trained via multi-agent reinforcement, which enabled them to learn solo navigation, obstacle avoidance, and coordinated firefighting.

  • In tests, this system achieved a 99.67% success rate in handling two fires simultaneously, with the robots splitting into teams to handle outbreaks.

Why it matters: Firefighting robots are already deployed on mine sites across Australia, but they still depend on human operators to function. This research marks an early but meaningful step toward swarms that can act autonomously — keeping human crews out of harm’s way when it matters most.

MIT

 🧠 MIT’s soft robots just got a brain

Image source: MIT / The National University of Singapore

The Rundown: MIT researchers just built an AI control system that lets soft robots learn a set of movements once, then adapt that knowledge to new tasks and unexpected disruptions in real time, with no retraining required.

The details:

  • The system lets a soft robotic arm learn a small library of basic movements once, then recombine and adapt them to new tasks without full retraining.

  • During operation, the plastic synapses update in real time, allowing the robot to respond to disturbances like unexpected pushes or shifting loads.

  • In tests, robots using this neural blueprint cut motion-tracking errors by up to about 50% and maintained stability even when actuators failed.

  • The framework generalizes across arm types and task categories — trajectory tracking, object placement, etc. — within a single unified system.

Why it matters: Soft robots that can learn a broad set of skills once and then adapt them to new situations make it far easier to deploy them outside tightly controlled lab or factory settings. That kind of human-like flexibility could unlock safer, more capable assistive, rehabilitation, and medical robots that work directly with people.

QUICK HITS

📰 Everything else in robotics today

New York Governor Kathy Hochul pulled a proposal that would have effectively legalized commercial robotaxi services in the state, stalling Waymo’s expansion there.

Unitree Robotics CEO Wang Xingxing said today’s humanoids are about as capable as a 10‑year‑old child, but large‑scale commercial deployment is still 3–5 years away.

An engineer trying to joystick-control his DJI Romo discovered a major security flaw that gave him access to live video and data from nearly 7K robot vacuums worldwide.

U.S.-based OpenMind is using its OM1 robot OS and marketing support to help Chinese robotics firms scale their hardware and expand sales into overseas markets.

Faraday Future co-CEO YT Jia told investors that FF’s EAI Robotics arm has cleared certifications and will start its first batch of deliveries next week.

Boston Dynamics’ Spot robot dog, upgraded with AI via the Orbit platform, now patrols factories to spot equipment issues early and prevent costly failure.

Uber still has a solid future in robotaxis, the Wall Street Journal argues, despite fears that autonomous vehicles will make its ride-hailing business obsolete.

U.S. researchers developed a real-time planning and control framework that lets bipedal robots quickly detect instability and adjust their steps, boosting fall recovery.

Infineon CEO Jochen Hanebeck said the German semiconductor giant is likely to profit from an expected boom in microchips for humanoids.

COMMUNITY

🎓 Highlights: News, Guides & Events

See you soon,

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

AI

What OpenAI and Jony Ive are building

Zach Mink • 7 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. We've known OpenAI and ex-Apple design guru Jony Ive have been building AI hardware since last May's $6.5B deal. What nobody knew was what the first device would actually be.

With new reporting revealing an upcoming smart speaker that can see, listen, and make purchases, the long-awaited collaboration is finally coming into focus — and it's heading straight for Amazon, Apple, and Google's turf.

In today’s AI rundown:

  • OpenAI's first AI device could be a smart speaker

  • The Rundown Roundtable: Our AI use cases

  • Self-host an n8n automation server in minutes

  • AI startup’s custom chip gives AI a 10x speed boost

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

OPENAI

🔊 OpenAI's first AI device could be a smart speaker

Image source: Lovart / The Rundown

The Rundown: OpenAI-Jony Ive's first hardware product will reportedly be a $200-$300 smart speaker with a built-in camera and facial recognition for purchases, according to The Information — backed by a 200+ team aiming to ship it by early 2027.

The details:

  • The team formed when OAI acquired Ive's startup Io Products for $6.5B in May, bringing in Apple veterans to lead hardware, design, and supply chain.

  • The speaker's camera will allegedly observe surroundings and “nudge (users) toward actions”, with a Face ID-like facial recognition feature for purchases.

  • AI-powered smart glasses are also planned, but won't hit production until at least 2028, with a smart lamp also created as a prototype.

  • OAI staffers have butted heads with LoveFrom over slow revisions and secrecy, with Ive's firm handling designs and the devices team working on the hardware.

Why it matters: OAI has never shipped a physical product, but the mystique surrounding Jony Ive has made its hardware a hotly anticipated launch. With Apple ramping up AI device plans and Amazon already rolling with Alexa+, OAI's window to define the category is shrinking fast — making the speaker a very important first swing.

TOGETHER WITH YOU.COM

📖 The handbook for your new AI agent employees

The Rundown You wouldn’t expect a new employee to know everything without onboarding, would you? The same concept goes for AI agents—and metadata is the key.

In this ebook, you’ll learn:

  • How metadata management drives AI success

  • Common pitfalls

  • The ROI of proper metadata management

Get the Ebook.

THE RUNDOWN ROUNDTABLE

💡The Rundown Roundtable: Our AI use cases

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 and daily lives.

Rowan, Founder & CEO: As a fast-moving startup, many of our team's best ideas come from random Slack threads, but get lost and never fully hashed out. Instead of spending hours a day manually adding tasks to our databases, we used Notion's new Agents feature (rolling out soon for GA) and built an "AI Project Manager" that monitors Slack messages daily and logs tasks autonomously.

Shubham, Editor: I used Gemini to analyze my blood work covering some 100 parameters. It instantly extracted and categorized all—from cholesterol to iron levels—into a clear, structured table, identified subtle deviations (such as slightly elevated eosinophils) and explained the clinical significance of each in plain language.

Finally, it translated these data points into a personalized wellness roadmap, suggesting natural lifestyle adjustments to optimize health markers.

Jennifer, Tech & Robotics Writer: Over the past year, I’ve relied heavily on LLMs to support my French citizenship application— including drafting required letters and tracking paperwork. Although I’m already fluent, I used ChatGPT to prep for the required language exam to get familiar with the test format for 100% confidence.

AI TRAINING

⚙️ Self-host an n8n automation server in minutes

The Rundown: In this guide, you will learn to set up your own n8n automation server — helping you run thousands of automations per month on n8n with just a $5 virtual server. Best of all, you can set it up in under 10 minutes.

Step-by-step:

  1. Go to railway.com/deploy/n8n, sign in with GitHub, and click Deploy Now. You don’t need to configure any variables. Click Deploy on the next screen

  2. After your automation server is deployed, click the n8n module, go to your custom link, and set up your email and password login

  3. Now, you can create a new automation by clicking Create workflow in your n8n dashboard — no need to set it up node-by-node

  4. Just ask Claude or ChatGPT to map out the automation you want as a JSON file. Then, in the new workflow, click the three dots, and import from file

Pro tip: You can invite users via email to your n8n server. This makes it great for client- or team-specific projects. You can even save API keys in the server for others to use.

For more tips on n8n + AI, check out our AI Automations Course.

PRESENTED BY OZ

⚡ Ship faster with agent-first engineering

The Rundown: Individual AI productivity gains hit a ceiling fast — without orchestration, they don't scale across your org, and leadership has no way to measure impact or enforce security standards. Oz is the platform built to change that.

Oz’s new report breaks down:

  • Why most companies fail at building their own agentic systems

  • How teams save hours per engineer per day using agent automations

  • What makes over 60% of agent-generated PRs actually achievable

Download the free report.

TAALAS

⚡️AI startup’s custom chip gives AI a 10x speed boost

Image source: Taalas

The Rundown: AI chip startup Taalas just emerged with HC1, a custom chip built to run a single AI model and nothing else — delivering responses roughly 100x faster than today's standard hardware and 10x the SOTA for extreme speed in outputs.

The details:

  • Taalas’ first chip permanently embeds Meta's Llama 3.1 8B model into the hardware rather than running it as software on general-purpose chips.

  • The result is near-instantaneous AI responses, with messages coming back in under 100 milliseconds at a fraction of the power and cost of other systems.

  • Llama 3.1 is small, older, and far from the frontier, but Taalas says it can retool chips for new models in just months — with a top-tier option planned by winter.

  • The startup pulled in $169M in new funding this round, bringing its total above $200M — with a mid-size reasoning model expected this spring.

Why it matters: The model baked into the first chip is far from competitive, but the tech itself is the story. The speed needs to be seen to be comprehended (demo here) — and if the approach scales to frontier models, it could change what's possible in areas like physical AI or agentic workflows where every millisecond matters.

QUICK HITS

🛠️ Trending AI Tools

  • 🗣️Unwrap Customer Intelligence - Turn unstructured customer feedback into data-backed insights that inform your product roadmap*

  • 💻 Claude in PPT - Anthropic's AI sidebar for building PowerPoint slides

  • 🎥 Replit Animation - Create professional animated videos from text prompts

  • ⚙️ Rork Max - Rork's AI-powered native iOS app builder

*Sponsored Listing

📰 Everything else in AI today

Sam Altman called concerns about ChatGPT's water usage "totally fake", arguing that creating AI may already be more energy-efficient than raising and ‘training’ a human.

Anthropic opened early access to Claude Code Security, a new tool that uses AI to detect hidden software vulnerabilities and suggest patches for human review.

Zyphra released ZUNA, an open-source AI trained on brain wave data that can clean up and reconstruct brain signals, an early step toward thought-to-text without surgery.

Pika Labs launched AI Selves, a new product that lets users create persistent AI clones that can post on social media, send messages, and interact across platforms.

Amazon’s Kiro AI coding agent reportedly caused a 13-hour AWS outage in December after autonomously deciding to delete and recreate an environment.

OpenAI’s Head of Codex posted he’s ‘beyond excited’ for the coming weeks, and that current coding agents will be seen as “so primitive that it will be funny in comparison.”

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 Gina T. in Minnesota:

"During a recent snowstorm, a big drift formed right behind my garage stall, and I couldn't get my car out. My husband was out of town, and I had never run the snowblower before. I went to the garage and took a picture of the back where all the controls were and loaded it into ChatGPT, and asked, "How do I start my snowblower?"

It identified the make and model, walked me through the steps on how to start the machine, and a few questions later, I was clearing my driveway!"

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

Zuck vs. Instagram addiction

Jennifer Mossalgue • 5 minutes

Read Online | Sign Up | Advertise

Good morning, tech enthusiasts. In a first-of-its-kind trial in Los Angeles, Mark Zuckerberg took the stand to defend against claims that Instagram was engineered like a ‘digital casino’ to hook kids.

With emails showing Zuck personally overruled child-safety and mental-health experts, the verdict could set the tone for thousands of similar lawsuits waiting in the wings.

In today’s tech rundown:

  • Zuck defends Instagram in landmark trial

  • Microsoft turns glass into a 10K-year hard drive

  • Feds charge 3 engineers in Google chip theft

  • Stanford’s new do-it-all respiratory vaccine

  • Quick hits on other tech news

LATEST DEVELOPMENTS

META

🎰 Zuck defends Instagram in landmark trial 

Image source: Ideogram / The Rundown

The Rundown: Mark Zuckerberg took the stand to defend Meta in a landmark LA jury trial over social media addiction, rejecting claims that Instagram was deliberately engineered to hook teens and harm their mental health. Plaintiffs argue otherwise.

The details:

  • The case centers on a 20-year-old woman who says compulsive use of Instagram and YouTube as a child fueled anxiety and suicidal depression.

  • Plaintiffs argue Instagram and YouTube were built as “digital casinos,” using addictive features like filters and infinite scroll to maximize time-on-platform.

  • Zuck maintained that Instagram offers positive value and that the plaintiff’s struggles stem from broader life problems, not product design.

  • Meta emails show Zuck personally overruled at least 18 mental health and child-safety experts who urged the company to curb beauty filters.

Why it matters: This is the first jury trial to test the wave of social media addiction lawsuits. Attorneys claim Meta and YouTube engineered features to hook “teens and tweens,” while shelving internal warnings about risks. Meanwhile, governments are already moving to restrict or ban social media access for under-16s.

MICROSOFT

💿 Microsoft turns glass into a 10K-year hard drive

Image source: Microsoft Project Silica

The Rundown: Microsoft has etched palm-sized slabs of ordinary glass into data “books” capable of storing 4.8 terabytes — the equivalent of roughly 2M books or 200 4K movies — and projecting their survival for at least 10K years.

The details:

  • The glass withstands heat, radiation, water, and demagnetization, making it virtually indestructible by the standards of conventional storage.

  • Inside Microsoft’s prototype archive, autonomous robot shuttles climb shelving units, retrieve the requested glass slab, and feed it to decoding systems.

  • The write speed, currently a few megabytes per second, targets potential customers including cloud providers, national archives, and media companies.

  • Because the glass “books” need no power, they could shrink archive footprints and dramatically cut the energy and hardware cost of storing cold data.

Why it matters: Project Silica could offer a way to lock humanity’s critical digital records into glass that outlasts every hard drive and cloud data center we own today. And unlike more exotic ideas such as DNA storage or 5D crystals, it builds on a familiar material — putting it closer to real-world deployment than most of its competitors.

GOOGLE

🧐 Feds charge 3 engineers in Google chip theft

Image source: Unsplash

The Rundown: U.S. prosecutors indicted three Silicon Valley engineers — including two former Google employees — for allegedly stealing hundreds of confidential files on Pixel processors and other proprietary chip designs.

The details:

  • A federal grand jury hit the trio, two Iranian-born sisters and one of their husbands, with 14 felony counts.

  • Prosecutors allege the sisters, while employed at Google, secretly copied hundreds of restricted files, including design data for the Tensor chip.

  • The trio allegedly moved the stolen documents to third-party messaging channels and personal devices, then on to contacts and storage in Iran.

  • All three were arrested in San Jose and made initial court appearances, and if convicted, could face prison sentences of up to 20 years.

Why it matters: The case lands just weeks after a separate conviction of another former Google engineer for stealing AI trade secrets, showing how aggressively U.S. authorities are now pursuing alleged tech espionage tied to strategic chip and AI technologies.

STANFORD

🦠 Stanford’s do-it-all respiratory vaccine 

Image source: Ideogram / The Rundown

The Rundown: Stanford researchers just developed an experimental intranasal vaccine that, in mice, defends the lungs for months against a range of viruses, bacteria, and even allergens — a potential leap toward a single shot that does it all.

The details:

  • Three doses of the vaccine, called GLA‑3M‑052‑LS+OVA, dramatically cut coronavirus levels in mouse lungs.

  • The vaccine’s broad coverage spans SARS‑CoV‑2 and related coronaviruses to drug-resistant hospital superbugs.

  • The secret is in the formula: a triple-adjuvant platform designed to rewire how the immune system responds at the source.

  • If it clears human trials, the technology could collapse today’s lineup of separate flu, COVID, and RSV shots into one annual nasal spray.

Why it matters: A single nasal spray guarding against the full spectrum of respiratory threats would fundamentally change vaccination and pandemic preparedness. Rather than endlessly reformulating shots to chase evolving strains, this platform aims to prime the immune system to handle whatever pathogen comes next.

QUICK HITS

📰 Everything else in tech today

Amazon just ended Walmart’s 13-year run to claim the No. 1 spot on the Fortune 500 for the first time.

SoftBank plans to spend $33B on a 9.2‑gigawatt natural‑gas power plant on the Ohio‑Kentucky border, potentially to feed data centers tied to its OpenAI partnership.

SpaceX Falcon 9 rocket reentry from a discarded upper stage left a plume of lithium and other metals in the Earth’s upper atmosphere, a new study found.

Tesla rolled out a new “entry” Cybertruck — a dual‑motor all‑wheel‑drive model starting around $60K — while cutting the high‑end Cyberbeast’s price by $15K.

Ford is developing a $30K midsize electric pickup for 2027, using an F1‑style skunkworks team and bounties to strip out weight, parts, and cost.

A new report finds that almost $1B in U.S. government research funding over the past decade flowed into projects involving Chinese labs, Bloomberg reported.

Thrive Capital closed a new $10B fund — its largest ever and nearly twice the size of its last — aimed at investments in companies such as OpenAI, Stripe, and SpaceX.

Rivian launched a native Apple Watch app that lets owners use their wrist as a digital key to lock or unlock the vehicle, open windows, and adjust cabin temperature.

Meta will shut down its standalone Messenger website in April and redirect users to Facebook’s messages page or the mobile app instead.

COMMUNITY

🎓 Highlights: News, Guides & Events

See you soon,

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

AI

The handshake refusal heard around the AI world

Zach Mink • 6 minutes

Read Online | Sign Up | Advertise

Good morning, AI enthusiasts. India rolled out the red carpet for the world's AI leaders this week, hoping for a show of unity. What it got instead was Sam Altman and Dario Amodei turning a photo op into an awkward viral moment.

The duo’s hand-hold dodge is peak tech-world absurdity, but underneath the meme is a rivalry between two of the most powerful AI labs that shows no sign of slowing down.

In today’s AI rundown:

  • The handshake refusal heard around the AI world

  • Google's Gemini 3.1 Pro doubles up on reasoning

  • Write viral YouTube scripts with NotebookLM

  • Corporate giant Accenture ties AI usage to promotions

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

OPENAI & ANTHROPIC

🤝 The handshake refusal heard around the AI world

Image source: India AI Impact Summit

The Rundown: OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei seemingly refused to hold hands during a group photo with Indian PM Narendra Modi at the India AI Impact Summit — a viral moment symbolizing the clash between the two AI giants.

The details:

  • Modi pulled tech leaders on stage for a hand-linked chain, lifting arms with Altman and Pichai — with Altman and Amodei awkwardly raising fists instead.

  • Altman later downplayed the moment, telling reporters he was "confused" when Modi grabbed his hand and he "wasn't sure what was happening.”

  • The moment follows Anthropic's Super Bowl ad campaign mocking OAI's decision to put ads in ChatGPT, which Altman called "clearly dishonest”.

  • OAI also hired the creator of AI agent OpenClaw last week, a potential source of contention given Anthropic’s issues with the original OpenClawd name.

Why it matters: If you’ve never seen the series ‘Silicon Valley’, it’s worth a watch to show how prescient it was on the ridiculousness of the tech world. While the moment makes for a viral meme, it also shows the state of affairs between top AI labs — far from the collaboration hoped for from leaders of the most important tech of our time.

TOGETHER WITH AIRIA

💼 Redefine enterprise AI with Airia

The Rundown: Airia delivers unified AI security, orchestration, and governance built for enterprises accelerating AI adoption. Deploy agents fast while maintaining control, bridging the gap between innovation and compliance.

Dive deeper with Airia’s 2026 State of AI Report, which reveals:

  • Key insights on enterprise AI adoption trends

  • Critical challenges and emerging opportunities ahead

  • A comprehensive guide for navigating AI transformation

Learn more.

GOOGLE

🧠 Google's Gemini 3.1 Pro doubles up on reasoning

Image source: Google

The Rundown: Google just released Gemini 3.1 Pro, bringing a massive reasoning upgrade, benchmark-topping performance, and overall SOTA capabilities while keeping API pricing identical to its predecessor.

The details:

  • 3.1 Pro scored 77.1% on the ARC-AGI-2 reasoning benchmark, up from 31.1% on Gemini 3 Pro and passing both Opus 4.6 (68.8%) and GPT-5.2 (52.9%).

  • The model also takes the top spot on benchmarks for science, competitive coding, MCP use, agentic search, and more.

  • Google positions 3.1 as the core intelligence behind last week's big Deep Think update, now available across the Gemini app, NotebookLM, and dev tools.

  • Pricing is identical to 3 Pro with the same 1M token context window, coming in cheaper than frontier model options from Anthropic and OpenAI.

Why it matters: After letting Anthropic and OpenAI control the headlines in 2026 so far, Google has answered back in a big way over the past few weeks — first with Deep Think and now sliding back into the ‘world’s top model’ conversation with an impressive 3.1 Pro launch. We expect a counter answer from OpenAI sooner rather than later.

AI TRAINING

✏️ Write viral YouTube scripts with NotebookLM

The Rundown: In this guide, you will learn to use NotebookLM to “watch” YouTube videos for you, then turn them into ideas and scripts for your own videos.

Step-by-step:

  1. Find a long-form YouTube video that you like the structure of. Go to NotebookLM > Create new > Add sources > Websites and paste the link in

  2. Once NotebookLM analyzes it, make sure it’s selected, and prompt: “Reverse engineer the structure of this video. Give me a video outline and 4 video concepts with the same structure for my brand, [brand]”

  3. Select the video concept you like best and instruct GoogleLM to generate a script for that idea. Click Save to note > click the three dots > Export to Docs

  4. Now you have a script in Google Docs that’s ready to record

Pro tip: You can save your best scripts as sources to give NotebookLM a better idea of what you like.

PRESENTED BY FIDDLER

🔎 The ultimate agentic observability guide

The Rundown: Fiddler AI’s Agentic Observability Guide breaks down what’s needed to monitor multi-agent systems so they succeed in production.

The guide covers:

  • Tracing AI failures across workflows

  • Leveraging the latest monitoring tools

  • Pinpointing the root cause of agent behavior

  • And more

Download the guide today and learn how to build better AI agent.

AI AT WORK

👀 Corporate giant Accenture ties AI usage to promotions

Image source: Lovart / The Rundown

The Rundown: Consulting giant Accenture is now reportedly monitoring weekly AI tool usage for senior employees and tying adoption directly to leadership promotions, in an attempt to bring veteran staff on board with the tech’s rising use in the workplace.

The details:

  • Three consulting execs told FT that getting senior partners to adopt AI is far harder than with junior staff, a firmwide seniority problem across the industry.

  • Associate directors aiming for promotions will now have AI tool logins tracked weekly, with usage flagged as a "visible input" to leadership reviews.

  • Accenture says 550K+ of its 780K staff have gone through AI training, though employees called the AI tools used in-house “broken slop generators”.

  • CEO Julie Sweet made headlines last year when she said on an earnings call that the firm would “exit” staff who don’t reskill for AI’s rise.

Why it matters: There is some irony in the senior employees being unwilling to adapt to the AI boom at the same time the tech is eating entry-level positions. But the wave of job transformation is here, and not learning AI will be a far more rapid equivalent of resisting the internet when it comes to competing and succeeding at work.

QUICK HITS

🛠️ Trending AI Tools

  • 🤖 Scrunch - See how AI interprets your site, run a free audit, and unlock the new way to reach customers*

  • 🧠 Gemini 3.1 Pro - Google's flagship model with SOTA reasoning gains

  • 📸 Pomelli - Google’s tool to turn product shots into marketing assets

  • 🚀 Claude Sonnet 4.6 - Anthropic’s mid-tier AI, rivaling Opus at 1/5 the cost

*Sponsored Listing

📰 Everything else in AI today

OpenAI is reportedly nearing a record $100B+ funding round backed by Amazon, SoftBank, Nvidia, and Microsoft, potentially lifting its valuation to over $850B.

Reddit is piloting an AI-powered shopping feature that converts community product recommendations into buyable carousels with pricing and retailer links.

ElevenLabs obtained the first-ever insurance policy covering AI voice agents, with its platform earning a certification that lets enterprises insure AI actions.

AMC Theatres is refusing to screen an AI-created film during its previews, pulling the contest-winning short from its preview lineup before the planned two-week run.

AI industrial startup Emanate launched out of stealth with autonomous revenue agents targeting the U.S.’ $5T industrial supply chain.

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 Nic C. in Edinburgh, Scotland:

"I am a teacher at an excellent online school. We use Canvas to deliver our curriculum. I have started to use Gemini to write code for canvas pages so that I can write a lesson in slides, share that with Gemini, and then let the AI build the HTML page.

What is really helpful is that the AI can take info from the presentation and make sure that I am producing a page that ticks all of the school’s prerequisites for Canvas pages or assignments based on the school rubrics."

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