Rowan Cheung(@rowancheung)

Source: https://t.co/93uwLKlnUB I break down stories like this every day in my free newsletter. Kee...

Rowan Cheung (@rowancheung) — Mon, 20 Apr 2026 15:13:35 GMT

Source: spectrum.ieee.org/boston-dynamic…

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬0🔄2❤️4👀6474📊1

⚡ Powered by xgo.ing

Boston Dynamics just gave its robot dog a brain that reasons about the physical world. Google DeepM...

Rowan Cheung (@rowancheung) — Mon, 20 Apr 2026 15:13:32 GMT

Boston Dynamics just gave its robot dog a brain that reasons about the physical world.

Google DeepMind's Gemini Robotics model is now running inside Spot, the four-legged robot already deployed at thousands of facilities worldwide.

The upgrade centers on something called embodied reasoning.

Basically, instead of following rigid instructions, Spot now interprets what it sees and makes decisions on its own.

One standout example: Spot can now look at a pressure gauge, zoom in on the needle, calculate the intervals between tick marks, and deliver an accurate reading. All autonomously.

The model also introduces:

> Multi-view reasoning, combining feeds from multiple cameras to understand its environment even when objects are blocked from one angle

> Built-in success detection, so Spot knows when it's failed and decides whether to retry or move on

That last part is a key requirement for true autonomy.

Most companies building AI-powered robots are still stuck in the demo phase. Boston Dynamics is one of the few with actual paying customers... which means thousands of Spots collecting real-world data that feeds back into making the AI smarter.

That flywheel is hard to compete with.

Your browser does not support the video tag.

🔗 View on Twitter

💬27🔄42❤️247👀36377📊64

⚡ Powered by xgo.ing

Research paper link: https://t.co/hU71XsNNUv I break down stories like this every day in my free ne...

Rowan Cheung (@rowancheung) — Fri, 17 Apr 2026 15:14:23 GMT

Research paper link: nature.com/articles/s4192…

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬0🔄0❤️8👀7037📊1

⚡ Powered by xgo.ing

MIT engineers built an AI wristband that controls robots by reading your hand muscles. It works by ...

Rowan Cheung (@rowancheung) — Fri, 17 Apr 2026 15:14:21 GMT

MIT engineers built an AI wristband that controls robots by reading your hand muscles.

It works by using an ultrasound to capture images of the muscles and tendons in your wrist.

An AI algorithm then translates those images into the exact position of all 5 fingers, tracking 22 different ways your hand can bend, twist, and extend.

In demos, wearers directed a robotic hand to play piano just by moving their own fingers.

To put the complexity in perspective, it takes:

> 34 muscles
> 27 joints
> 100+ tendons
...Just to scroll your phone

This device captures it all from a band the size of a smartwatch.

They tested across 8 volunteers with different hand sizes, tracking everything from sign language letters to holding scissors.

The team is now collecting data from hundreds more users to train robots in surgical procedures.

Something like this could one day become the universal remote for robots and virtual worlds, and we're way closer than most people expect.

Your browser does not support the video tag.

🔗 View on Twitter

💬15🔄13❤️88👀25794📊28

⚡ Powered by xgo.ing

Research paper link: https://t.co/E0cbG7laMZ I break down stories like this every day in my free ne...

Rowan Cheung (@rowancheung) — Tue, 14 Apr 2026 15:12:38 GMT

Research paper link: pnas.org/doi/10.1073/pn…

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬1🔄2❤️8👀10699📊2

⚡ Powered by xgo.ing

This robot can be torn apart and still keeps moving. Northwestern researchers built modular robot l...

Rowan Cheung (@rowancheung) — Tue, 14 Apr 2026 15:12:35 GMT

This robot can be torn apart and still keeps moving.

Northwestern researchers built modular robot legs that each operate as fully independent machines:

> Own battery
> Own motor
> Own sensors
> Own brain

A single module can jump, roll, and turn on its own.

Bolt a few together and they become entirely new species of robot that can walk, flip, and sprint across outdoor terrain.

Some move like lizards, others drag themselves forward like seals.

Every behavior was learned in simulation and transferred to the real world with zero fine-tuning.

The goal is to mass-produce these parts so anyone can snap them together like Lego bricks, building robots on-site in disaster zones or hazardous environments, customized to whatever the situation needs.

Wild!

Your browser does not support the video tag.

🔗 View on Twitter

💬16🔄20❤️126👀40363📊38

⚡ Powered by xgo.ing

Research paper link: https://t.co/XPbDNPcU6m I break down stories like this every day in my free ne...

Rowan Cheung (@rowancheung) — Mon, 13 Apr 2026 15:13:38 GMT

Research paper link: cell.com/cell/fulltext/…

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬2🔄4❤️11👀9243📊5

⚡ Powered by xgo.ing

Microsoft's AI can now detect cancer from a $10 tissue sample. For context, every time tumor cells ...

Rowan Cheung (@rowancheung) — Mon, 13 Apr 2026 15:13:35 GMT

Microsoft's AI can now detect cancer from a $10 tissue sample.

For context, every time tumor cells are tested, doctors create a basic microscope slide to study tissue up close.

These slides show cell shapes and structures, but they can't reveal which immune cells are actually fighting the cancer.

That deeper picture is critical for knowing if a patient will respond to immunotherapy... But the advanced imaging needed costs THOUSANDS.

So Microsoft built GigaTIME -- an AI system that generates advanced imaging from the cheap slides hospitals already collect.

The system was trained on 40 million cancer cells, then applied to over 14,000 patients across 51 hospitals spanning 24 cancer types.

The AI found over 1,200 hidden connections between immune cell behavior and tumor growth that researchers couldn't find before... because the data simply didn't exist at this scale.

When validated against 10,000 additional patients from a completely separate database, the results held up.

The model is now open source, so any hospital worldwide can use it on samples they already have.

I think this is one of the most impactful AI papers I've seen this year!

Your browser does not support the video tag.

🔗 View on Twitter

💬20🔄102❤️405👀40428📊114

⚡ Powered by xgo.ing

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Thu, 09 Apr 2026 15:17:33 GMT

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬1🔄1❤️13👀7870📊3

⚡ Powered by xgo.ing

Meta's new AI can predict your brain better than a brain scan. TRIBE v2 is a foundation model train...

Rowan Cheung (@rowancheung) — Thu, 09 Apr 2026 15:17:30 GMT

Meta's new AI can predict your brain better than a brain scan.

TRIBE v2 is a foundation model trained on 1,000+ hours of brain imaging data from 720 people.

You feed it a video, sound clip, or text, and it predicts:

> Which brain regions light up
> How strongly
> And in what order

When tested on people it had never seen, the model's predictions were actually more accurate than most real brain scans (which get distorted by heartbeats, breathing, and movement).

Researchers then replicated decades of classic neuroscience experiments entirely inside the software.

No scanner, no human subjects.

The model correctly identified the brain's face recognition center, language network, and emotional processing regions on its own.

But look at what else Meta has been building:

> Ray-Ban smart glasses that see and hear what you do
> A wristband that reads nerve signals
> And now a model that predicts how your brain responds to any piece of content

There's no evidence these are all connected... but Meta now has a complete picture of attention, from the stimulus to the neural response.

Your browser does not support the video tag.

🔗 View on Twitter

💬38🔄31❤️179👀28199📊69

⚡ Powered by xgo.ing

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Wed, 08 Apr 2026 15:17:01 GMT

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬2🔄3❤️6👀7357📊3

⚡ Powered by xgo.ing

Halter now manages nearly 650,000 cows, and is currently raising at a $2B valuation. Half the world...

Rowan Cheung (@rowancheung) — Wed, 08 Apr 2026 15:16:58 GMT

Halter now manages nearly 650,000 cows, and is currently raising at a $2B valuation.

Half the world's habitable land is farmland. This is AI being applied to one of the oldest industries on earth and I think we're going to see a lot more of this.

Your browser does not support the video tag.

🔗 View on Twitter

💬5🔄2❤️25👀8957📊9

⚡ Powered by xgo.ing

An AI cow collar just created a billion-dollar company. Farmers draw boundaries on a phone app, and...

Rowan Cheung (@rowancheung) — Wed, 08 Apr 2026 15:16:37 GMT

An AI cow collar just created a billion-dollar company.

Farmers draw boundaries on a phone app, and the collars guide cows using sound and vibration.

It works by collecting over 6,000 data points per min, feeding ML models that track grazing patterns, predict disease, and detect peak breeding times.

And they call it the "Cowgorithm."

It saves farmers 20-40 hours of labor per week and has already replaced over 800,000 km of physical fencing.

The collars are also solar-powered and built with bulletproof glass to survive daily wear from 1,000-pound animals.

The founder grew up on a New Zealand dairy farm where his parents started work at 4am and often worked 100+ hours a week. He built Halter to solve the problems he watched growing up.

Your browser does not support the video tag.

🔗 View on Twitter

💬40🔄52❤️357👀55484📊108

⚡ Powered by xgo.ing

Github link: https://t.co/vZ7ygd0IVT

Rowan Cheung (@rowancheung) — Tue, 07 Apr 2026 13:35:58 GMT

Github link: void-model.github.io

💬2🔄0❤️12👀7883📊4

⚡ Powered by xgo.ing

Researchers at Netflix just released a new AI model It erases objects from video, then rewrites the...

Rowan Cheung (@rowancheung) — Tue, 07 Apr 2026 13:26:23 GMT

Researchers at Netflix just released a new AI model

It erases objects from video, then rewrites the physics of the entire scene as if that object never existed

It's called VOID (Video Object and Interaction Deletion)

Current inpainting tools simply paint over the gap left by an object's removal, but they ignore the chain of events that would inevitably follow.

For example:
> If you remove a car from a crash scene with a basic AI tool, and the other car still looks like it collides with something invisible
> But VOID understands that once the car is gone, the other one would keep moving forward without impact

The output is still rough and limited to a few seconds of footage, but with Netflix's entire content library to train on... this tech has a very clear path to improve.

Your browser does not support the video tag.

🔗 View on Twitter

💬38🔄22❤️173👀45929📊70

⚡ Powered by xgo.ing

All predictions are now live on Flood Hub for free: https://t.co/OQhcCP7ubp I break down stories li...

Rowan Cheung (@rowancheung) — Mon, 06 Apr 2026 15:13:19 GMT

All predictions are now live on Flood Hub for free: sites.research.google/floods/l/30.26…

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬1🔄0❤️11👀7608📊3

⚡ Powered by xgo.ing

Google's new AI can predict flash floods 24 hours before they strike. How it works: > Uses Gemini ...

Rowan Cheung (@rowancheung) — Mon, 06 Apr 2026 15:13:17 GMT

Google's new AI can predict flash floods 24 hours before they strike.

How it works:

> Uses Gemini to extract confirmed flood locations and times from global news

> Builds a dataset of past events that never formally existed.

> That dataset feeds a neural network

> The neural network combines real-time weather forecasts with local terrain, soil absorption rates, and urban density

> Can flag at-risk areas with results that match the accuracy of America's National Weather Service

> Can do this in countries that have almost no flood monitoring infrastructure at all

Flash floods kill more than 5,000 people every year, but most victims never see them coming.

A 12-hour warning alone can reduce flash flood damage by 60%.

...For billions of people without any early warning system, this could be the difference between life and death.

Your browser does not support the video tag.

🔗 View on Twitter

💬45🔄27❤️157👀21923📊67

⚡ Powered by xgo.ing

Source: https://t.co/5mjehWMRgI I break down stories like this every day in my free newsletter. Kee...

Rowan Cheung (@rowancheung) — Fri, 03 Apr 2026 15:15:32 GMT

Source: zzk273.github.io/LATENT/

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬4🔄1❤️17👀8065📊6

⚡ Powered by xgo.ing

Researchers just taught a robot to play tennis. From just clips of a few amateur players performing...

Rowan Cheung (@rowancheung) — Fri, 03 Apr 2026 15:15:29 GMT

Researchers just taught a robot to play tennis.

From just clips of a few amateur players performing basic forehands, backhands, and shuffles...

...a robot learned one of the fastest, most coordinated physical skills there is.

Insane!

Your browser does not support the video tag.

🔗 View on Twitter

💬42🔄17❤️137👀38543📊57

⚡ Powered by xgo.ing

Source: https://t.co/JUwoUsmGQx I break down stories like this every day in my free newsletter. Kee...

Rowan Cheung (@rowancheung) — Fri, 20 Mar 2026 16:15:22 GMT

Source: technologyreview.com/2026/03/10/113…

I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬11🔄5❤️15👀10758📊12

⚡ Powered by xgo.ing

Every Pokémon Go player unknowingly trained future robots. Over the years, players photographed 30 ...

Rowan Cheung (@rowancheung) — Fri, 20 Mar 2026 16:15:20 GMT

Every Pokémon Go player unknowingly trained future robots.

Over the years, players photographed 30 billion images from cities around the world.

And now that data is teaching delivery robots to navigate streets.

Niantic (company behind Pokémon Go), spun out an AI company called Niantic Spatial that's turning years of player data into a visual positioning system for machines.

When 500 million people installed the game and pointed their phones at buildings, landmarks, and street corners, they were also capturing detailed location data that normal GPS can't match:

> GPS signals bounce off skyscrapers and lose accuracy by up to 50 meters in dense cities.

> That's a dealbreaker for sidewalk delivery robots that need to stop at exactly the right point.

> Niantic Spatial trained a model on those 30 billion images to pinpoint a robot's exact location based on what its cameras see, not satellite signals.

> Delivery startup Coco Robotics is the first partner, using the system to guide thousands of robots across LA, Chicago, and Miami.

Don't think anyone thought they were training robots when they were chasing Pikachu during COVID.

Your browser does not support the video tag.

🔗 View on Twitter

💬67🔄24❤️151👀36689📊83

⚡ Powered by xgo.ing

More info @orcahand if you want to learn more, or even get one for yourself :-) https://t.co/6GpH9yH...

Rowan Cheung (@rowancheung) — Thu, 19 Mar 2026 16:12:23 GMT

More info @orcahand if you want to learn more, or even get one for yourself :-) orcahand.com/hardware

P.S. I break down stories like this every day in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

💬5🔄6❤️33👀15574📊13

⚡ Powered by xgo.ing

This robotic hand can be 3D printed by anyone and assembled in under 8 hours. Researchers at ETH Zu...

Rowan Cheung (@rowancheung) — Thu, 19 Mar 2026 16:12:20 GMT

This robotic hand can be 3D printed by anyone and assembled in under 8 hours.

Researchers at ETH Zurich created the Orca hand, fully open-sourced with artificial bones and tendons.

For context, advanced robotic hands cost over $100,000 and require constant maintenance...

Orca costs under $2,000. 50x less (!)

A self-calibration system maps every motor to every joint, eliminating the manual tuning that tendon-driven hands usually need.

Each fingertip has built-in tactile sensors covered by silicone skin.

The hand can actually feel when it touches something, giving it feedback to grip objects without crushing them or letting them slip.

It can hold over 20 lbs, learn tasks by watching human demonstrations, and transfer skills trained in simulation directly to the real world.

The team proved its durability by having it pick up and place a cube over 2,000 times across 7 hours with no human intervention.

The full design files and source code are open source, so any robotics lab in the world can start building one today.

Your browser does not support the video tag.

🔗 View on Twitter

💬111🔄419❤️3000👀211730📊833

⚡ Powered by xgo.ing

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Tue, 17 Mar 2026 16:13:23 GMT

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

Source: eu-startups.com/2026/02/with-o…

💬2🔄1❤️15👀9238📊4

⚡ Powered by xgo.ing

This robotic hand can weave itself together in minutes. Allonic developed a process that "braids" r...

Rowan Cheung (@rowancheung) — Tue, 17 Mar 2026 16:13:20 GMT

This robotic hand can weave itself together in minutes.

Allonic developed a process that "braids" robot bodies around a 3D-printed skeleton in a single automated step.

The tech draws from the textile industry, using braided fibers instead of traditional mechanical joints and bearings.

Their braiding system grows tendon-driven structures directly onto skeletal cores, creating flexible yet durable parts without screws or assembly.

One finger only needs 4 skeletal elements. Then thousands of braided fibers wrap around them to handle force and articulation.

The software converts designs directly into machine code that runs the braiding process.

But what matters is that robotic hands like this can cost up to $30,000, ~20-30% of a humanoid robot's total price.

Allonic's braiding process could crush those costs while enabling custom designs at scale.

Your browser does not support the video tag.

🔗 View on Twitter

💬73🔄199❤️940👀73747📊263

⚡ Powered by xgo.ing

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Mon, 16 Mar 2026 16:16:08 GMT

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

Source: theguardian.com/uk-news/2026/m…

💬0🔄0❤️9👀16158📊1

⚡ Powered by xgo.ing

A surgeon just removed a man's prostate ...and he was 1,500 miles away. From an office in London, D...

Rowan Cheung (@rowancheung) — Mon, 16 Mar 2026 16:16:05 GMT

A surgeon just removed a man's prostate ...and he was 1,500 miles away.

From an office in London, Dr. Prokar Dasgupta controlled a 4-armed robot at a hospital in Spain.

The robot was fitted with a 3D camera and connected to London via fiber optic cable with a backup 5G connection (lag was just 0.06 seconds).

The story is wild:

> 62-year-old Paul Buxton had expected to fly to England and wait months on the NHS list.

> Instead, the surgeon came to him... through a fiber optic cable.

> Dasgupta is already scheduled to perform the procedure again on March 14th, with around 20,000 surgeons watching live.

Access to care is going to look so different in the future.

For context, in many low-income countries, there are fewer than 1 trained surgeon per 100,000 people.

The US alone is projected to face a shortage of up to 20,000 surgical specialists by 2036.

...Remote surgery doesn't just make medical care more convenient, it may be the only way some patients ever receive it one day.

💬39🔄36❤️233👀30614📊71

⚡ Powered by xgo.ing

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Thu, 12 Mar 2026 16:11:26 GMT

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

Source: nature.com/articles/s4146…

💬2🔄3❤️14👀9748📊5

⚡ Powered by xgo.ing

South Korean researchers built a paper-thin robot that squeezes through 3mm gaps and lifts 70x (!) i...

Rowan Cheung (@rowancheung) — Thu, 12 Mar 2026 16:11:23 GMT

South Korean researchers built a paper-thin robot that squeezes through 3mm gaps and lifts 70x (!) its weight.

The flexible robotic "sheet" mimicks myosin, the protein that powers muscle contractions in your body.

Inside the sheet are dozens of microscopic air chambers stacked in layers.

When air flows through them in a specific sequence, tiny feet move like muscles pushing and returning.

Each foot repeats a simple push-and-return motion, allowing the robot to travel long distances with small movements.

In demos, it navigated through a pipe with just 3 millimeters of clearance while carrying a camera 23x heavier than itself.

The team also attached it to a standard 2-finger gripper, instantly giving it the ability to rotate and reposition objects mid-air without moving the arm.

One day this could be used in minimally invasive surgery, allowing robots to navigate through the body's smallest passages.

Your browser does not support the video tag.

🔗 View on Twitter

💬25🔄81❤️372👀52530📊99

⚡ Powered by xgo.ing

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Wed, 11 Mar 2026 16:15:21 GMT

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

Source: SWARM Biotactics linkedin.com/posts/s-wilhel…

💬3🔄7❤️55👀35158📊17

⚡ Powered by xgo.ing

NATO is testing live cockroaches as AI-powered spy drones. Incredible AI engineering, but also some...

Rowan Cheung (@rowancheung) — Wed, 11 Mar 2026 16:15:18 GMT

NATO is testing live cockroaches as AI-powered spy drones.

Incredible AI engineering, but also something I kinda wish I hadn't learned about:

> Swarm Bio-tactics wired real cockroaches with electronic backpacks containing AI hardware, radios, cameras, and microphones.

> Cockroaches are steered by sending electrical signals directly into the insect's nervous system

> They can crawl through rubble, tunnels, and spaces where drones can't fly, and troops shouldn't go, transmitting data back the entire time.

> Within one year, they went from concept to field-validated systems with paying NATO customers, including the German military.

The qualities that make them useful for military recon (small, silent, nearly undetectable) are exactly what make them creepy.

...International laws weren't written with cyborg insects in mind.

Your browser does not support the video tag.

🔗 View on Twitter

💬188🔄429❤️1920👀2291126📊678

⚡ Powered by xgo.ing

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Roboti...

Rowan Cheung (@rowancheung) — Tue, 10 Mar 2026 16:13:21 GMT

I break down stories like every day this in my free newsletter. Keep up with the latest in AI/Robotics in 5 min a day: therundown.ai/subscribe

Source: wired.com/story/say-good…

💬0🔄0❤️12👀15574📊2

⚡ Powered by xgo.ing

Fun fact: The first transatlantic internet cable is being pulled off the ocean floor right now. Alm...

Rowan Cheung (@rowancheung) — Tue, 10 Mar 2026 16:13:18 GMT

Fun fact: The first transatlantic internet cable is being pulled off the ocean floor right now.

Almost no one knows it's happening.

TAT-8 went live in 1988. First fiber-optic cable to connect Europe and the US.

Isaac Asimov called it a "maiden voyage across the sea on a beam of light."

The engineers behind it genuinely believed it might be the last transatlantic cable the world would ever need. The tech was so far ahead of copper they thought global bandwidth was solved.

But it filled to capacity within 18 months.

That forced the industry to immediately start laying more cables, a race between demand and infrastructure that still hasn't stopped.

After developing a fault too expensive to fix, the cable was retired in 2002. It's been sitting on the seabed ever since.

Now a small crew aboard this ship is dragging it back up.

Meanwhile, Meta announced Project Waterworth last year, a 50,000 kilometer cable stretching across 5 continents, partly to feed the data demands of AI.

...The entire global internet runs on cables like this, built and maintained by a few thousand people working in places most of us will never see.

Your browser does not support the video tag.

🔗 View on Twitter

💬51🔄294❤️2362👀453316📊451

⚡ Powered by xgo.ing

Source: https://t.co/cu13Hyi74f Did a video breakdown on this as well:

Rowan Cheung (@rowancheung) — Sun, 08 Mar 2026 17:32:39 GMT

Source: nature.com/articles/s4158…

Did a video breakdown on this as well:

Your browser does not support the video tag.

🔗 View on Twitter

💬3🔄12❤️80👀19420📊19

⚡ Powered by xgo.ing

Microsoft found out how to etch 5 terabytes of data into a piece of glass ...and it survives for 10...

Rowan Cheung (@rowancheung) — Sun, 08 Mar 2026 17:29:10 GMT

Microsoft found out how to etch 5 terabytes of data into a piece of glass

...and it survives for 10,000 years!

> Project Silica is a storage system that uses ultrafast lasers to write data directly into glass.

> Each laser pulse lasts one trillionth of a second and carves microscopic structures called voxels into 301 layers inside ordinary glass (the same kind your food container is probably made out of).

> To read the data back, a microscope captures images of each layer, and an AI image recognition model decodes the patterns with zero errors.

But what's wild is the glass needs NO power to maintain its data. It's immune to heat, water, radiation, and magnetic fields that would normally wipe conventional hard drives over the years.

Accelerated aging tests predict the data will survive past 10,000 years at room temperature, twice as long as the oldest known writing in the world.

For context, today's cloud archives burn enormous amounts of energy just keeping data alive on degrading magnetic tape. Glass storage needs none of that.

💬95🔄184❤️1136👀67508📊275

⚡ Powered by xgo.ing

Exclusive from @theinformation by @steph_palazzolo https://t.co/rN0mTVtBll

Rowan Cheung (@rowancheung) — Thu, 18 Dec 2025 19:26:36 GMT

Exclusive from @theinformation by @steph_palazzolox.com/steph_palazzol…

Stephanie Palazzolo@steph_palazzolo

New: Brett Adcock, the CEO of robotics startup Figure AI, is starting a new AI lab called Hark, which will be funded by $100M of his personal capital. The startup aims to build "human-centric AI" and Adcock will also remain in his position at Figure.

theinformation.com/briefings/excl…

🔗 View Quoted Tweet

💬31🔄5❤️46👀30342📊24

⚡ Powered by xgo.ing

New AI lab just dropped. Brett Adcock (founder of Figure, the humanoid robotics company valued at $...

Rowan Cheung (@rowancheung) — Thu, 18 Dec 2025 19:25:43 GMT

New AI lab just dropped.

Brett Adcock (founder of Figure, the humanoid robotics company valued at $39B) is reportedly self-funding $100M into the new lab called "Hark."

The lab is building "human-centric AI" that can "think proactively, recursively improve and care deeply about people" 👀

💬120🔄30❤️297👀45806📊141

⚡ Powered by xgo.ing

It's clearer to me than ever that AEO is the next big distribution opportunity. Huge thanks to @dha...

Rowan Cheung (@rowancheung) — Tue, 16 Dec 2025 18:59:16 GMT

It's clearer to me than ever that AEO is the next big distribution opportunity.

Huge thanks to @dharmesh for taking the time and for the fantastic conversation!

Link to the full video on YouTube: youtu.be/fKjgKr-kgAY

💬10🔄0❤️52👀28866📊15

⚡ Powered by xgo.ing

Being early to SEO helped Dharmesh Shah build HubSpot into a $30B company. Now he's obsessed with s...

Rowan Cheung (@rowancheung) — Tue, 16 Dec 2025 18:50:20 GMT

Being early to SEO helped Dharmesh Shah build HubSpot into a $30B company.

Now he's obsessed with something new: AI ranking.

It's how to show up when people ask ChatGPT for recommendations instead of Googling.

I sat down with him to break down the entire playbook:

00 Intro
53 The AI ranking opportunity (AEO)
14 How to rank in AI chatbots
11 Using AI to adapt content for AEO
24 Training teams on "AI Intuition"
45 Content moats in the AI-slop era
57 Dharmesh's new AI agent startup
33 AI agent monetization strategy
38 The bottlenecks to autonomous agents
15 Building teams: humans vs agents
49 Why Dharmesh still codes as a billionaire

Your browser does not support the video tag.

🔗 View on Twitter

💬124🔄153❤️923👀210811📊451

⚡ Powered by xgo.ing

Image source: https://t.co/Q45pkhoWDo

Rowan Cheung (@rowancheung) — Mon, 01 Dec 2025 16:13:56 GMT

Image source: reddit.com/r/ChatGPT/comm…

💬2🔄0❤️19👀15749📊4

⚡ Powered by xgo.ing

Adding to document where we currently are: "somewhere in the messy middle": > AI is mainstream (an...

Rowan Cheung (@rowancheung) — Mon, 01 Dec 2025 16:11:52 GMT

Adding to document where we currently are:

"somewhere in the messy middle":

> AI is mainstream (anyone on the street knows ChatGPT)
> But NOT impacting most real economic work significantly yet
> A small % of people have figured how to leverage intelligence to (quite literally) do 10-100x more work
> Majority of users are on free plans, using Chat like Google Search, but still benefiting

💬8🔄2❤️33👀18740📊11

⚡ Powered by xgo.ing

ChatGPT just turned 3 years old. 0 to 800M weekly users in that time. Fastest in history. Wild how...

Rowan Cheung (@rowancheung) — Mon, 01 Dec 2025 16:04:24 GMT

ChatGPT just turned 3 years old.

0 to 800M weekly users in that time. Fastest in history.

Wild how much the way we work, search, and learn has changed in such a little time.

💬66🔄24❤️181👀28723📊83

⚡ Powered by xgo.ing

https://t.co/O1IJ7LhlFP

Rowan Cheung (@rowancheung) — Wed, 26 Nov 2025 15:37:50 GMT

youtube.com/watch?v=UwoflM…

💬4🔄2❤️43👀20243📊18

⚡ Powered by xgo.ing

Over 3 years, I bootstrapped The Rundown to 2M+ subscribers and millions in revenue. But I never re...

Rowan Cheung (@rowancheung) — Wed, 26 Nov 2025 15:37:19 GMT

Over 3 years, I bootstrapped The Rundown to 2M+ subscribers and millions in revenue.

But I never really shared how.

To be honest, I have massive imposter syndrome.

I'm a college dropout, self-taught in everything. I never followed typical "business" advice.

But some friends convinced me that sharing what's actually working right now might help others who are building (warning: lots of unconventional hacks).

So I'm breaking out of my bubble a little bit.

Starting to document the behind-the-scenes as the business continues to grow.

My interview with EO just dropped, where I share:

- The original organic growth strategy for The Rundown
- Why we DON'T use AI to write (even as an AI newsletter)
- The AI workflows that save me 20+ hours/week
- How I landed interviews with Zuck and Satya

This was uncomfortable. But if even one person finds value in the weird stuff that worked for me, it's worth it.

Let me know what you think and if you found it valuable!

Link below for the full thing.

Your browser does not support the video tag.

🔗 View on Twitter

💬118🔄31❤️294👀93965📊163

⚡ Powered by xgo.ing

It's always an honor speaking with @demishassabis! In the full interview, we chat more on: -Proact...

Rowan Cheung (@rowancheung) — Tue, 18 Nov 2025 16:23:21 GMT

It's always an honor speaking with @demishassabis!

In the full interview, we chat more on:

-Proactive AI healthcare
-"Learning how to learn"
-Secret internal models + research

Listen in on Apple Podcasts: podcasts.apple.com/us/podcast/the…

Listen in on Spotify: open.spotify.com/show/2zQpIc96g…

💬6🔄4❤️51👀17427📊17

⚡ Powered by xgo.ing

Demis' dream digital coworker vision: an AI assistant "useful every moment of your life" for both wo...

Rowan Cheung (@rowancheung) — Tue, 18 Nov 2025 16:08:52 GMT

Demis' dream digital coworker vision: an AI assistant "useful every moment of your life" for both work and leisure.

It follows you across all devices - computer, phone, even smart glasses - understanding the physical world around you through multimodal capabilities.

Your browser does not support the video tag.

🔗 View on Twitter

💬1🔄5❤️64👀21832📊10

⚡ Powered by xgo.ing

Google just released Gemini 3, its most intelligent AI model yet. I caught up with Demis Hassabis, ...

Rowan Cheung (@rowancheung) — Tue, 18 Nov 2025 16:08:50 GMT

Google just released Gemini 3, its most intelligent AI model yet.

I caught up with Demis Hassabis, CEO of Google DeepMind, to ask about:

-Gemini 3 and Google’s AI strategy
-Google's new Antigravity tool
-Medical-grade AI

Here's what he said:

💬137🔄59❤️805👀157103📊240

⚡ Powered by xgo.ing

What's the weirdest way you've made money with ChatGPT (directly or indirectly)? Genuinely curious ...

Rowan Cheung (@rowancheung) — Wed, 12 Nov 2025 18:00:44 GMT

What's the weirdest way you've made money with ChatGPT (directly or indirectly)?

Genuinely curious about the non-obvious stuff.

💬7🔄0❤️15👀3690📊9

⚡ Powered by xgo.ing

Pretty blown away by demand. If you DM'd or emailed me - I see you! But there are too many to reply...

Rowan Cheung (@rowancheung) — Tue, 11 Nov 2025 13:00:49 GMT

Pretty blown away by demand.

If you DM'd or emailed me - I see you! But there are too many to reply individually.

So if you want a shot at the role, please apply through the link above so I can review everything in one spot 👆

💬7🔄1❤️21👀14475📊8

⚡ Powered by xgo.ing

Fun fact: In the early days of building @therundownai, I was building an AI notetaker as a side proj...

Rowan Cheung (@rowancheung) — Tue, 11 Nov 2025 12:22:06 GMT

Fun fact: In the early days of building @therundownai, I was building an AI notetaker as a side project

(Still own the domain noat.ai)

The only major competitors back then were Otter and Fireflies.

Crazy to see where they ended up.

💬12🔄2❤️307👀101330📊40