Recent breakthroughs in generative AI have centered largely on language and imageryfrom chatbots that compose sonnets and analyze text to voice models that mimic human speech and tools that transform prompts into vivid artwork. But global chip giant Nvidia is now making a bolder claim: the next chapter of AI is about systems that take action in high-stakes, real-world scenarios.
At the recent International Conference on Learning Representations (ICLR 2025) in Singapore, Nvidia unveiled more than 70 research papers showcasing advances in AI systems designed to perform complex tasks beyond the digital realm.
Driving this shift are agentic and foundational AI models. Nvidias latest research highlights how combining these models can influence the physical worldspanning adaptive robotics, protein design, and real-time reconstruction of dynamic environments for autonomous vehicles. As demand for AI grows across industries, Nvidia is positioning itself as a core infrastructure provider powering this new era of intelligent action.
Bryan Catanzaro, vice president of applied deep learning research at Nvidia, described the companys new direction as a full-stack AI initiative.
We aim to accelerate every level of the computing stack to amplify the impact and utility of AI across industries, he tells Fast Company. For AI to be truly useful, it must evolve beyond traditional applications and engage meaningfully with real-world use cases. That means building systems capable of reasoning, decision-making, and interacting with the real-world environment to solve practical problems.
Among the research presented, four models stood outone of the most promising being Skill Reuse via Skill Adaptation (SRSA).
This AI framework enables robots to handle unfamiliar tasks without retraining from scratcha longstanding hurdle in robotics. While most robotic AI systems have focused on basic tasks like picking up objects, more complex jobs such as precision assembly on factory lines remain difficult. Nvidias SRSA model aims to overcome that challenge by leveraging a library of previously learned skills to help robots adapt more quickly.
“When faced with a new challenge, the SRSA approach analyzes which existing skill is most similar to the new task, then adapts and extends it as a foundation for learning, Catanzaro says. This brings us a significant step closer to achieving generalization across tasks, something that’s crucial for making robots more flexible and useful in the real world.
To make accurate predictions, the system considers object shapes, movements, and expert strategies for similar tasks. According to one research paper, SRSA improved success rates on unseen tasks by 19% and required 2.4 times fewer training samples than existing methods.
Over time, we expect this kind of self-reflective, adaptive learning to be transformative for industries like manufacturing, logistics, and disaster responsefields where environments are dynamic and robots need to quickly adapt without extensive retraining,” Catanzaro says.
Biotech breakthroughs
The biotech sector has traditionally lagged in adopting cutting-edge AI, hindered by data scarcity and the opaque nature of many algorithms. Protein design, essential to drug development, is often hampered by proprietary data silos that slow progress and stifle innovation.
To address this, Nvidia introduced Proteínaa large-scale generative model for designing entirely new protein backbones. Built using a powerful class of generative models, it can produce longer, more diverse, and functional proteinsup to 800 amino acids in length. Nvidia claims it outperforms models like Google DeepMinds Genie 2 and Generate Biomedicines Chroma, especially in generating large-chain proteins.
According to a paper on Proteína, the team trained the model using 21 million high-quality synthetic protein structures and improved learning thanks to new guidance strategies that ensure realistic outputs during generation. This breakthrough could transform enzyme engineering (and, by extension, vaccine development) by enabling researchers to design novel molecules beyond what occurs in nature.
“What makes it especially powerful is its ability to generate proteins with specific shapes and properties, guided by structural labels, Catanzaro says. This gives scientists an unprecedented level of control over the design processallowing them to create entirely new molecules tailored for specific purposes, like new medicines or advanced materials.
A new AI tool for autonomous vehicles
Another standout from ICLR 2025 is Spatio-Temporal Occupancy Reconstruction Machine (STORM), an AI model capable of reconstructing dynamic 3D environmentslike city streets or forest trailsin under 200 milliseconds. With minimal video input, it produces detailed, real-time spatial maps that can inform rapid machine decision-making. Nvidia sees STORM as a tool for autonomous vehicles, drones, and augmented reality systems navigating complex, moving environments.
“One of the biggest backlogs in current models is that they often rely heavily on optimizationan iterative process that takes time to refine and produce accurate 3D reconstructions,” says Catanzaro. “STORM tackles this by achieving high-accuracy results in a single pass, significantly speeding up the process without sacrificing quality.
STORMs potential extends beyond vehicles. Catanzaro envisions applications in consumer tech, such as AR glasses capable of mapping a live sports game in real timeallowing viewers to experience the event as if they were on the field. STORMs real-time environmental intelligence moves us closer to a future where machines and devices can perceive, understand, and interact with the physical world as fluidly as humans do, he says.
While STORM is built to help machines understand the physical world in real time, Nvidia is also pushing the boundaries of how large language models reasonthrough a project called Nemotron-MIND. This 138-billion-token synthetic pretraining data set is designed to enhance both mathematical and general reasoning. At its core is MIND, a new framework that turns raw math-heavy web documents into rich, multi-turn conversations that mirror how humans work through problems together.
By turning dense math documents into conversations between people with different levels of understanding, MIND helps AI models break down problems step by step and explain them naturally. This method doesnt just teach models what the right answer isit helps them learn how to think through problems like a person would.
According to its research paper, a seven-billion-parameter model trained on just four billion tokens of MIND-style dialogue outperformed much larger models trained on traditional data sets. It showed significant gains on key reasoning benchmarks like GSM8K (grade school math), MATH, and MMLU (massive multitask language understanding), and achieved a 2.5 percent boost in general reasoning when integrated intoan LLM.
Can startups and researchers keep up?
Training and deploying advanced AI models requires substantial GPU resources, often out of reach for smaller players. To close this gap, Nvidia is rolling out its next-gen AI models through Nvidia Inference Microservices (NIMs), a suite of containerized, cloud-native tools designed to simplify deployment across different infrastructures. NIM includes prebuilt inference engines for a wide array of models, helping organizations integrate and scale AI with fewer computing resources.
Improving efficiency has always been a major focus for us, Catanzaro says. Ultimately, our goal is to democratize access to AI capabilities and make deployment practical at every scale, regardless of their computing resources, to harness the power of AI.”
As agentic and foundational AI becomes more capable and more embodied, the future of tech may hinge on how effectively it works with humans. Its critical to identify and support use cases across diverse fields, Catanzaro says.
A little over two years ago, AI avatars took the internet by storm as people flocked to apps like Lensa, which generated idealized, often fantastical portraits of themselves. But in the ever-elusive offline world, another, quieter trend has been bubbling up: real portraits, made by real people.
Portrait commissions have been on the rise. In 2024, the Royal Society of Portrait Painters, headquartered in London, saw a 40% increase in portrait commissions from American clients now make up roughly 20% of their total. The U.S. has a fascination with the Royal Family more than we do sometimes, says Martina Merelli, fine art commissions manager at the Royal Society of Portrait Painters. Its an acknowledgment of the quality of work.
It’s no wonder Americans are fascinated. Since its founding in 1891, the Royal Society of Portrait Painters, also known as RP, has been the society of choice for the British Royal Familys public and private commissions. Its members have famously painted the late Queen Elizabeth II, Princess Diana, the Duke of Edinburgh, the Prince and Princess of Wales, and Prince Harry. Notable figures like Dame Judy Dench, Sir David Attenborough, and Stephen Hawking have also been captured on canvas.
[Image: Courtesy of Ben Sullivan/the Royal Society of Portrait Painters]
But the commission service isn’t limited to the elite. As long as you have disposable income (a head and shoulder begins at around $6,500) anyone can commission a portrait. At a time where AI is squashing many artists, this particular art form is enduringperhaps as a symbol of our need for tangible human connection.
A brief history of portraiture
Portraits, like art more broadly, have long been seen as a mirror to society. Before the camera was invented, the only way to record someone’s likeness was to paint, or sculpt, a portrait of them.
But portraits were never just a recordthey were signifiers of wealth, taste, and power. In ancient Egypt, painted portraits were placed over mummies to guide them into the afterlife. In Ancient Rome, they were used to commemorate the dead and assert lineage. Emperors used them to reinforce authority. Dictators turned them into propaganda tools.
One of the first portraits to depict a merchant couple from the middle ranks of society appeared during the Renaissance, when the focus expanded from rulers, nobility, and clergy, to wealthy merchants, bankers, and scholars.
Today, portraiture remains intertwined with global politics and economic tides. Its no secret that many of our clients are brokers, bankers, hedge fund managerspeople whose decisions are deeply affected by how the market is going, Merelli says.
In 2024, the U.K. saw two major elections. These ushered in a transition from a conservative to a Labour government that directly impacted the tax structures around private schooling. Merelli recalls one acquaintance with three daughters in private schools remarking that taxation money used to be their art money.
The new faces of portraiture
[Image: Courtesy of Frances Bell/the Royal Society of Portrait Painters]
Despite its exclusive history, over the past few decades, the art of portraiture has become more accessible. Frances Bell, an RP member who has been painting portraits for over 20 years, says her clientele now includes newlyweds, young professionals, and parents wanting to leave behind a tangible heirloom. Its a time stamp, she says. Something important they will carry on.
Institutional portraits of CEOs, lawyers, chancellors, and the like still account for a big portion of the market. (Bell has also painted members of the royal family but these are cloaked in NDAs.) She believes the impulse behind a portrait commission often goes deeper than vanity. “I’m not saying it’s not there, I think it’s there for all of us, but I get people who want a little thrum of the life force to be put on into the canvas to last forever,” she says. “It’s that feeling of posterity, and permanence.”
Unsurprisingly, that kind of posterity doesnt come cheap. Merelliwho often acts as “cupid” between prospective clients and painters at the RPsays the average price for a portrait in 2025 has decreased from what it used to be, but it still hovers around $13,000$20,000. “You can go up to $130,000 depending who the artist is, what brief you have, but a comfortable number is probably $66,000 to $80,000 if you want a full length of yourself with your house in the background and the dogs.” (Frances, who was trained at the prestigious Charles. H. Cecil Studios in Florence, charges $10,000 and upwards for a head and shoulders paintingor about $4,000 for a charcoal drawing.)
A proud antithesis to AI
That portraiture remains popular is both a rejection of the zeitgeist and, paradoxically, a natural extension of it. It is a slow process that can take countless hours over many sittings, and that i precisely why it is appealing. “It’s quite confessional,” says Bell, who places great importance on the in-person sittings. “I have their secrets coming out of my ears.”
Everyone interviewed for this story emphasized the intimacy of the sitting process. Something about two people breathing the same air, in the same room, and looking at each other for hours. For Anthony Connolly, president of the Royal Society of Portrait Painters, this dynamic even shapes the vocabulary painters use. While photographs shoot, painters find a presence, come to a lightness. “You’re there, with your model and it’s like a triangular conversation, where the third point of the triangle is the thing you’re making,” says Connolly.
The connection goes both ways. For the painter, its an act of seeing. For the sitter, its an act of being seen. Its a bonding experiencean art formthat no algorithm can ever replicate.
An investment piece
Claudia Fisher, an American who moved to the UK around the beginning of the pandemic, was not allowed to divulge the cost of her paintinga head and shoulder by painter Paul Brason. Having never owned a piece of art before, the cost was “one giant gulp,” she says. But she has no regrets.
[Image: Courtesy of Paul Brason/the Royal Society of Portrait Painters]
Fisher, now 69, was reading a book about the social history of tiaras when I called her. After a multifaceted career as an opera singer and a classical architecture designer, she has turned to fashion and today runs a fashion label called Belle Brummell, which makes luxury jackets inspired by 18th and 19th century British couture.
Fisher wanted her portrait to act as a marketing tool for her designs. She had just wrapped up the first prototype of her jacket, when it dawned on her: what better way to evoke the historical spirit of her brand than to be portrayed in one of her own designs, in a composition reminiscent of the era? “I’ve always loved the idea of getting a portrait done because I had vision of myself being in a gorgeous dress,” she says with a laugh. “It wasn’t about immortalizing me, I just wanted a pretty dress.” She got a pretty jacket instead.
Fisher made four separate trips to Bath, where Brason lives, on four separate occasions. Brason also traveled to her and her husband’s house in Brighton to get a sense of her personality at home, take reference photos, and do a pencil sketch. The two are still in touch. “If were in the area Ill call and see if hes around,” she says. “These relationships continue.”
When the electric car startup Rivian was set to release its first vehicle in late 2021, the company made the unconventional choice. Instead of a more conventional neutral tone, it manufactured a significant amount of its initial production run in a custom color the company called Launch Green. It was a decision that ran counter to almost every color trend and automotive industry sales report, and one that’s come to shape the way the company builds out one of the most unique color palettes in the car business.
“Everybody buys black, white, or gray. Pretty much every single brand, they’re going to have that. And it doesn’t matter if you’re in the U.S., you’re in China, you’re in Europe, that’s what it is,” says Jeff Hammoud, Rivian’s chief design officer. “Those are the ones that people order. But they’re not the ones that create the most buzz or excitement.”
Launch Green, marketed as a limited run for the company’s R1T truck, bucked the trends and rose near the top of color rankings among Rivian fans and buyers. The Rivian forum on Reddit had such a heated debate over Launch Green’s merit that its moderator pinned a note to the top of the comments thread stating that it had been reported by some users for “incorrectly” placing Launch Green in second place. “I understand that many of you feel personally victimized by Launch Green not being #1. I encourage you to take a break from the internet or talk to a loved one for support,” the note read. Though the company doesn’t break down its sales figures publicly, Launch Green was immediately popular. Despite being a limited run, customers still ask for it nearly five years later.
[Photo: Rivian]
Colors that look good dirty
Considering the approach validated, the company has since put an uncommon amount of effort into its color palette, not only creating unique custom colors but also making those colors an extension of Rivian’s adventure-centric, California-inspired brand. From L.A. Silver to El Cap Granite to Red Canyon to Storm Blue, Rivian’s paint options purposely lean into an outdoorsy theme. The company just announced another limited edition paint and trim package, California Dune, a pale sand color that evokes off-roading in the desert.
[Image: Rivian]
“We wanted something that like looked crisp and clean and premium,” Hammoud says. “If it’s dirty you can’t really tell. It’s not this car that you feel like you constantly have to clean, like a black car.”
Rivian does offer its vehicles in black”Midnight” in the company’s parlancebut only reluctantly. Hammoud says Rivian CEO RJ Scaringe doesn’t like black, which he finds to be “not an optimistic color” and one that’s not exactly on brand. “But,” Hammoud says, “customers love it.”
For some, Rivian’s colors may just look like slightly different versions of a blue or red that any other car company might use. But according to Hammoud, Rivian’s colors have been carefully developed to reflect a spirit of adventure, while also being bespoke to the brand. “We want it to have that warmth that our brand has, and also something that invites you to get it dirty,” he says.
[Photo: Rivian]
That approach to color has become so ingrained in the brand’s approach that Rivian hired its own in-house paint specialist, enabling it to develop new color options faster. Even so, adding new colors to the palettethere have been 12 so farrequires a significant investment of time and coordination with suppliers. “The fascias, the mirrors, the door handles, parts of the liftgates, none of those are actually painted at our plant,” Hammoud says. “So we have to work with all these different suppliers to essentially take that same color and make sure it matches identical.”
[Photo: Rivian]
Adding a new color can take years, but Hammoud says that limited color runs can happen much faster, since the company’s manufacturing facility in Normal, Illinois, can swap a color into the production line for a short time before returning to a more standard color. Bringing a new color like California Dune into the lineup for a limited run is another way for the company to generate some brand buzz. “It’s a fun and I wouldn’t say easy but a light lift for us to be able to go and add freshness to the vehicle by offering a new color,” Hammoud says.
[Photo: Rivian]
Rivian is also careful about when to take a color out of the lineup. One discontinued color, Compass Yellow, had consistently high Net Promoter Scores, a measure of how likely a customer is to recommend a product to others. “People were the most passionate about that color and Red Canyon, which are really low take rates for us,” Hammoud says. Though the yellow was dropped from the lineup, the red is still available.
[Photo: Rivian]
These color choices are partly driven by sales figures and customer demand, but Hammoud says the company’s overall approach to color is more closely tied to the adventurous image it’s trying to create with its off-road-ready truck and SUV models. The company pays attention to color trends in the automotive world, but isn’t concerned with simply keeping pace with competitors. “Everything we do from a color standpoint is influenced by the types of products that we think align with our brand, align with our customers. And a lot of that starts from outdoor adventure gear, footwear, backpacks,” Hammoud says. This extends to other sides of Rivian design, like the brand’s distinctive headlights, which were inspired by a rock climbing carabiner.
[Photo: Rivian]
But Hammoud says color may be one of the most important elements of Rivian’s vehicles. “Color is a big part of purchase consideration for people,” he says. Ultimately the cars are products, and the company is trying to sell them. Color, he argues, helps make the cars more distinctive, which leads to more customer interest, and maybe a foothold in a crowded marketplace.
“Finding inspiration from outside of automotive is a big part of it,” Hammoud says. “If you don’t do that, you’re just going to feel like you’re every other car brand.”
My worst workday habit is that I’m a compulsive web page checker.
Throughout the day, I’m constantly refreshing the same handful of sites for updates. I’ll check the metrics on my newsletters, swing through a subreddit or two, and click through some tech news sitesand that’s before even getting to email and social media. Every time I do this, it’s hard to refocus.
So I was pretty eager to try Aloha Browser’s new “Snips” feature, which uses AI to periodically monitor web pages and notify you when things change. I figured that by having AI check web pages on my behalf, I could avoid the urge to do so myself and be better at staying on task.
It’s helped at least a little, but both Aloha and I still have some work to do.
[Image: Aloha]
How Snips works
Snips is currently available in the desktop version of Aloha for Mac and Windows, appearing as a little box-and-scissors icon next to the address bar. Clicking the icon brings up a selector tool for highlighting the part of the page you want to keep track of.
[Screenshot: Jared Newman]
After selecting a snippet, you’ll see a menu for setting up alerts. Choose how often Aloha should check for updates (the default is once per day, but you can go as frequently as every five minutes), then write a sentence describing what changes it should watch for. For instance, if you wanted to monitor the price on a product page, you could write something like “notify me when the price falls below $300.”
[Screenshot: Jared Newman]
In my case, I’ve set up a handful of Snips to cut down on compulsive page checking:
For the pages where I check on newsletter metrics, I’ve instructed Aloha to only notify me when certain parameters change.
I like to check the New York Yankees subreddit, so I’ve asked Aloha to notify me when new posts are created.
If I post on social media, I can create a temporary Snip that alerts me if the responses reach a certain threshold.
I have alerts set up for when new stories appear on Techmeme, just to make sure I don’t miss anything important.
For email, I have Aloha alert me of replies to existing conversation threads.
Behind the scenes, Aloha uses on-device AI to analyze page content, then takes routine snapshots of the page to see if things change. For the notification requests, it uses a mix of on-device AI processing and large language models from Grok and OpenAI, but Aloha says no browsing data leaves your device in most cases. (The browser does send some especially complex tasks to a remote server for processing, but requires permission first and deletes the data immediately after.)
Once you’ve created some Snips, they’ll appear as screenshots on Aloha’s new tab page. You can tweak the notifications from here, but you can also shuffle and resize the screenshots into a kind of glanceable information dashboard.
Why it makes sense
There are plenty of other ways to monitor information online. I use CamelCamelCamel for price alerts on Amazon, for instance, and you can always turn on push notifications for email and social media.
[Image: Aloha]
But Aloha’s Snips feature is a useful alternative because of how granular it can get. You can set up price alerts on any retail site without sharing your contact information, and you can limit social media notifications to specific types of responses or reactions. The alerts come through the Mac or Windows notification tray, so your email inbox and phone notifications stay uncluttered.
Room for improvement
That’s not to say Aloha’s Snips feature is perfect. It’s subject to the same vagaries as other generative AI tools, which means things may not always work as expected. For instance, I’ve experienced some instances of false positive notifications when nothing changes, or repeat notifications for things I’ve been alerted to already.
Aloha’s page refresh capabilities also don’t seem to work 100% of the time. One snippet I set up for the “Newest” section on Techmeme refused to update, and Aloha showed error messages while trying to update standard Reddit pages. (As a work-around, I had to create a snippet on old.reddit.com instead.) If the information you need requires extra clicking or scrolling after reloading the page, it’s not going to work with Snips either.
[Photo: Aloha]
And even when things are working properly, I still have to provide the appropriate degree of willpower. I don’t need Aloha to check Reddit every five minutes, but if I set the interval to be too infrequent, I’ll likely get antsy and start checking it myself. That’s entirely a me problem.
Aloha is not my main browser, and it was not really on my radar until the Snips feature arrived. It’s made by a small team based in Cyprus, and touts an emphasis on privacy, but I still prefer the power-user features in the likes of Vivaldi and Floorp.
Even so, it’s easy enough to keep running in the background to discourage my compulsive checking habit. I’m going to keep doing that to take a little of the weight off my mind.
An empty light truck is cruising along a sun-drenched highway of Qionghai, a city in Hainan Island, the southernmost part of China. As the car that’s filming overtakes it, we can see the truck has no driver. In fact, it doesnt even have a cabin: Its front is just a flat wall crowned by what looks to be sensors and cameras. Its an eerie and surreal view, a Headless Horseman of trucks just as scary as an actual headless horseman.
View this post on Instagram A post shared by Ai Revolution (@aitherevolution)
The futuristic yet cheap-looking vehicle is part of a fleet of driverless light trucks that can carry 1,000 parcels each completely unattended over a range of more than 110 miles. These vehicles, operated by Chinese logistics giant ZTO Express, are the vanguard of a silent, state-sponsored effort to revolutionize the way China ships goods around the country. Their fleet is already vastly outperforming the efforts of startups in the U.S. They navigate Hainans suburban and rurl routes thanks to an artificial intelligencepowered computer that sees the world in 3D using lasers and high-resolution cameras. The trucks are capable of obeying traffic lights, dodging obstacles, yielding to pedestrians, and “talking” to the road itself and other vehicles.
The program began in November 2024 with a single vehicle, followed by three additional trucks as part of a pilot overseen by the Eastern Postal Administration of Hainan Province. Its director Zhang Zhi called it at its launch the beginning of a new intelligent era for the regions courier industry. The initial pilot focused on Qionghais campuses, commercial districts, and residential areas, but ZTO quickly expanded it throughout the island and the rest of China.
Beijing turbocharge
It’s just another step in Chinas road to automated logistics. The company already had experience with this autonomous technology for last-mile and long-haul logistics. In July 2024, ZTO launched autonomous delivery vans in Taizhousouth of Shanghaieach capable of carrying 600 to 800 parcels per tripdouble the capacity of human couriers. These vans, which started development in 2021, are equipped with 360-degree cameras and AI-trained obstacle detection. They now handle nearly a third of last-mile deliveries in Taizhous industrial zones. The vehicles use V2X (vehicle-to-everything) communication systems, a technology that allows them to talk to traffic lights, road sensors, and other vehicles in real time. Allegedly, V2X reduces collisions and optimizes traffic flow by sharing data like speed, direction, and road conditions.
Then, in August, ZTO deployed 400 autonomous heavy-duty trucks across Chinas highway network, developed jointly with Shanghai-based autonomous driving startup Inceptio Technology and Dongfeng Commercial Vehicle, a subsidiary of Chinas state-owned Dongfeng Motor Group. This marked the largest single delivery of intelligent freight trucks globally at the time, each equipped with light detection and ranging sensors that create 3D maps of surroundings, redundant braking systems, and Inceptio autonomous driving software, a proprietary system designed to optimize long-haul freight efficiency by reducing fuel consumption and human error. The company claims the software has driven 124 million miles, a truly impressive record.
The key to this success is Beijings aggressive push to fully automate its logistics sector as part of its ambitious 2030 agenda, a national program aimed at building a modern, harmonious, and creative society, according to the World Bank. Hainan was an experiment in which regulatory agility paved the way for this rapid scaling. The province slashed certification requirements to just 1,864 miles of testing, compared to Chinas most populous provinceGuangdongwhich has a 9,32018,600-mile mandate.
Since then, 12 provinces have adopted Hainans fast-track certification model, and Beijing has allocated $1.4 billion to retrofit highways with 5G networks and V2X infrastructure. 5Gs ultralow latency provides near-instant data transmission and it ensures autonomous vehicles can process sensor data and communicate with infrastructure without delays, which is a prerequisite for safe operation at high speeds. ZTOs own proprietary unmanned vehicle management platform, launched a year ago, now monitors a 200 autonomous vehicles fleet across 40 cities, tracking everything from battery levels to pedestrian interactions in real time.
An army of ghost trucks and bots
And its all scaling up this year. As of April 2025, 27 driverless vehicles operate at the companys Laiwu logistics park in Shandong, south of Beijing. Their routes synced with workers handheld scanners. Government officials in this province confirm plans to deploy at least 1,500 such vehicles across Shandong by late 2025, targeting a 50% reduction in labor costs. This shift is driven by necessity: labor costs in Chinas logistics sector have risen 8% annually since 2022, while e-commerce parcel volumes exceeded 130 billion in 2024. Thats why Beijing is so adamant to make this happen.
ZTO is not alone in this. Alibaba’s logistics arm Cainiao claims to have deployed “thousands” of autonomous delivery robots and vehicles during its 2024 Cainiao Smart Global Logistics Summit. Chinese retail giant JDs logistics division has 600 autonomous vehicles in operation, making millions of deliveries. And food delivery titan Meituan has been deploying hundreds of fully driverless delivery vehicles in major urban centers like Beijing and Shenzhen, according to its Q3 2024 earnings call. Neolix has been deploying thousands of its homegrown autonomous vehicles for various commercial delivery applications since 2021.
Its a stark contrast with whats happening in the U.S., where theres a patchwork of state- and city-led policies. Companies like Kodiak Robotics and Gatik are testing autonomous trucks for middle-mile delivery, with Gatik operating a small fleet of box trucks for Walmart in Arkansas. However, deployments remain constrained by fragmented state regulations and a lack of centralized infrastructure investment. For example, California requires permits from both the DMV and Public Utilities Commission for commercial autonomous operations, while Texas allows driver-out testing in the state: In May 2024, Pittsburgh-based autonomous truck technology company Aurora Innovation announced that its first commercial trucksdeveloped with Volvoare now driving between Dallas and Houston. The company said that, to date, its self-driving tech has completed 1,200 miles without a driver. Compare that to Inceptio’s 124 million miles.
A for smaller vehicles, Waymo Via’s publicly acknowledged deployment of fully driverless delivery vans is likely in the dozens, primarily within pilot programs. Waymo-powered trucks were in trial runs until 2022, but the company stopped its efforts in 2023. Nuro claims it has expanded its autonomous vehicle operations in a handful of cities: Mountain View, Palo Alto, and Houston. Notably, Amazon has not disclosed large-scale deployment numbers for fully driverless road vehicles in commercial operation, and its nutty air delivery system is just not flying as Jeff Bezos probably expected.
By 2030, S&P Global Mobility estimates that China will dominate autonomous freight, with 250,000 Level 4 logistics vehicles in operation, compared to 230,000 in the U.S.most of which will remain focused on ride-hailing, not freight. Seems optimistic for the U.S. side. China’s strong push for automated driving, bolstered by significant government support and regulatory frameworks, positions it as a potential leader in the development of autonomous vehicle technology and relative to commercialization of the autonomous vehicle industry, the report says. A centralized strategy, which has resulted in 28,000 miles of roads that are now open for autonomous vehicles with 16,000 licenses issued nationwide. Only time will tell if the U.S. can overtake Beijing, but for now, I can only see a formidable army of Chinese ghost trucks amassing beyond the Great Wall and shaping the future of roads, while we are still playing with cars in geofenced Disneyland rides.
This fall, NASA scientist Kris Bedka flew into Hurricane Helene to test a device that uses lasers to create ultradetailed wind measurements. It could be the key to unlocking better storm predictions.
The new device is called the Aerosol Wind Profiler (AWP), and its been in the works at NASA for about four years. The AWP uses the Doppler effect to create real-time, 3D maps of wind patterns above the Earths surfacedata that, before now, has been much more difficult to capture.
Bedka is the AWPs principal investigator at NASAs Langley Research Center, and he has spent more than 100 hours in the air testing the device in collaboration with the National Oceanic and Atmospheric Administration (NOAA), which runs the National Weather Service (NWS). He believes the AWP could be the missing link in weather agencies abilities to accurately map severe weather events like hurricanes and thunderstorms.
[Photo: NASA/David C. Bowman]
The shortcomings to current wind data collection
To create predictions for severe weather phenomena, agencies like the NWS collate a vast swath of data including atmospheric temperature, moisture levels, and pressure patterns, typically pulled from satellite readings. Wind patterns, both on the ground and above ground level, are another piece of assembling the overall puzzle.
But when it comes to fitting wind patterns into the model, there are a few limitations. If forecasters need wind measurements close to the Earth, Bedka says, its fairly simple to take readings using sensors that can be mounted on the ground. But what’s most important for weather forecasting is having a sense of the three-dimensional picture of the windso winds not just at the ground, but many miles above us, which all combine to drive the weather that we experience at the ground, he says.
Currently, to get a sense of the 3D wind picture, forecasters mainly use weather balloons. These balloons give accurate readings, Bedka says, but there are only around 1,300 launch sites across the globe, meaning their data is fairly limited. Another tool, called geostationary satellites, can use snapshots of cloud cover and atmospheric moisture patterns to calculate wind vectors, but only at the cloud top, meaning that the 3D wind picture is still missing. Many experts believe that tools like the AWP are the “missing link” to address this problem.
[Photo: NASA/David C. Bowman]
How the AWP uses lasers to make a 3D wind map
Before making a detailed 3D wind map, scientists need to understand two main factors: how fast wind currents are moving and in what direction. The AWP does that by tracking the movement of particulatesincluding tiny pieces of cloud matter, dust, smoke, pollution, and sea salt that are all floating in the atmosphereto see how wind is buffeting them at a given moment in time.
To capture the movement of those particulates, the AWP is mounted to an aircraft with viewing ports underneath it. From there, the instrument emits 200 pulses of laser energy per second toward the atmosphere in two opposite directions, where they scatter and reflect off the particulate matter. This scattering causes a measurable change in the laser pulse wavelength, also known as the Doppler effect.
You’ve probably heard of the Doppler effect before, and youve experienced it yourself, Bedka says. You hear an ambulance coming towards you, and at one particular distance, it sounds very high pitched, and then as it comes by you and then goes away from you, you hear the pitch changethat’s due to the Doppler effect. A Doppler wind lidar kind of behaves in an analogous way.
In simple terms, the altered frequencies of laser light that bounce back from particulates give the AWP the information needed to calculate wind speed and direction, even measuring conditions at different altitudes in the atmosphere simultaneously. All of those details can then be stitched together to create a complete 3D wind map.
[Photo: NASA/Maurice Cross]
The AWP flies through Hurricane Helene
In 2022, Bedka says, NOAA solicited new technologies for accurate wind measurement, which had been an ongoing challege for the agency when trying to predict severe weather. Since Bedkas team had just wrapped up their AWP prototype, they proposed an aircraft flight campaign that would validate the tools effectiveness.
NOAA agreed to fund the proposal, and last fall, Bedka took flight for over 100 hours in a kind of flying laboratory, installed inside a 1970s-era DC-8 aircraft. The lab came outfitted with NASAs AWP and its High-Altitude Lidar Observatory (HALO), another tool built to measure water vapor, aerosols, and cloud properties. Over the course of the flight campaign, AWP and HALO worked together to create ultradetailed 3D maps of wind patterns and aerosol layers. Bedka was aiming to collect data from as wide a range of weather conditions as possibleand, as it happened, that included Hurricane Helene.
Because Helene was a relatively well-predicted storm, Bedkas team had time to plan a flight route that would allow the AWP to measure as close to the storm center and the highest winds that were available to us. Given the planes limited six-hour flight range, Bedka and the crew flew through the edges of the hurricane in several legs on September 26, traveling down the western edge of the storm, going around the eye in the Gulf of Mexico, and heading back up the East Coast. In all, it took about nine hours.
Bedka, who has flown in several NASA aircrafts through intense thunderstorms, says the conditions were choppy but not too severe. During the hurricane flight, his team was able to collect a rich database of wind measurements that proved the AWPs potential effectiveness during severe weather.
Whats next for the AWP
For now, the AWP is just in testing phases, but NASA is currently working to make it more widely available.
That would involve bringing on an agency or commercial partner, like NOAA, willing to invest further in the technologyideally, by adapting it for use on smaller satellites rather than flying it up on planes. Currently, the AWP is about the size of a coffee table, but to fit on a vessel set for space, researchers would need to shrink it to about one-tenth its current size, Bedka says. (According to a NASA spokesperson, the AWP project hasn’t been impacted by federal budget and staff cuts at NASA and NOAA.)
Ideally, NASA would be able to create a constellation of AWPs orbiting the Earth that could measure winds simultaneously all across the globe. With such a wide swath of data, prediction models for extreme weather would become significantly more accurate.
Severe storms don’t just pop up just out of the clear blue sky on a random day, Bedka says. They form because all the ingredients align in order to make them become as intense as they are. What we’re trying to do with this technology is to measure the winds with as much spatial and vertical detail as permitted by laser technology. We’ve already found that when this data is put into weather prediction models, it has a really big impact.
In 2021, I quit a 15-year career as a tech executive in the finance industry and pursued content marketing and journalism.
When I tell this story, Im often met with, You did what? People cant wrap their heads around such an unexpected career shift.
While I quit my tech job, it wasnt an overnight decision. In fact, it was something Id been considering for a long time.
{"blockType":"creator-network-promo","data":{"mediaUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/04\/workbetter-logo.png","headline":"Work Better","description":"Thoughts on the future of work, career pivots, and why work shouldn't suck, by Anna Burgess Yang. To learn more visit workbetter.media.","substackDomain":"https:\/\/www.workbetter.media","colorTheme":"blue","redirectUrl":""}}
A career pivot is much different than simply finding a new job in the same industry. If youve contemplated the same, youve probably asked yourself, Can I do this? Will it be worth the change? What if it doesnt work out?
As someone whos pivoted not once, but twice, in the past four years (first, I changed industries, and then I became self-employed), I can tell you yes, its worth it. But you need to be ready for the impact such a change will have on your career. Heres how to tell if a major career change is right for you.
You cant stop thinking about it
The first sign that youre ready for a career pivot is that you cant stop thinking about it. You dream not only of finding a new job, but doing something completely different.
You start to explore new skills you might need to acquire, or how to prepare your résumé so its attractive to hiring managers in a new field.
Once you start thinking nonstop about a career change, those feelings never go away. Trust me. Theyll only expand over time.
You start to plan
Your dreams become more concrete when you start to consider a timeline. For me, it was, Can I make this change within five years? The more I thought about it, the shorter my timeline became. Staying in my current job instead of pursuing something different became unbearable.
Youll start to think about a financial plan. A career pivot may involve a lower salary (at least in the beginning) if youre starting in a new industry with little experience.
Suddenly, my five-year timeline turned into six months. Then I quit without another job lined up (though I was at the end stages of several interviews and an offer came in a few days later).
Youre comfortable with unknowns
Changing careers can be uncomfortable, especially once you start a new job. The saying You dont know what you dont know is particularly apt. All the research in the world cant fully prepare you for your first day doing something completely different.
As a longtime employee at my previous job, I went from being the most knowledgeable person in the room to one of the least. I went from being the boss to having a boss. I had no idea how long it would take me to catch on to the nuances of my new career. There were times I felt inadequate due to my lack of knowledge.
But I also knew that my previous careerthough far differenthad given me skills others in my peer group didnt have. I understood contract negotiation, project management, and resource planning. I leaned into those, which helped me move past my discomfort.
Youre willing to pivot again (if needed)
As you plan to pivot to a new career, you have to ask yourself, What if it doesnt work out?
Maybe you end up hating the work. Maybe moving from a senior role to a junior role bothers you too much. Maybe its not the work itself, but you land at a company that isnt a good fit, and now have to continue the job search.
Im here to tell you: Its okay to pivot again. Trashing your résumé is an outdated concept. After I pivoted to a new career and joined a marketing agency, I left within eight months and joined a different agency. One year after that, I pivoted again. Now Im self-employed.
As it turns out, that itch I felt to try something different was really a drive to strike out on my own. I didnt know it at the time. I had to make a few stops along the way to figure out exactly what I wanted to do.
What you think you want and what you actually want might be two different things. But chances are, if youre willing to make the leap once, youll be willing to do it again.
Anna Burgess Yang is part of the Creator Network at Fast Company, covering topics like work culture and the intersection of technology and work (including the impacts of AI).
Anna is a former tech executive who spent more than 15 years at a financial technology company, including roles as a product manager and the director of customer success. In 2021, she joined the Great Resignation and quit her job to pursue content marketing and journalism. Today, she is a freelance writer who works with prominent B2B SaaS companies, specializing in fintech, automation, and AI. Additionally, her work has appeared in The Chicago Tribune, The Chronicle of Higher Education, and several banking publications.
You can learn more about Anna at annabyang.com. You can also follow her on LinkedIn, Threads, and Bluesky.
{"blockType":"creator-network-promo","data":{"mediaUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/04\/workbetter-logo.png","headline":"Work Better","description":"Thoughts on the future of work, career pivots, and why work shouldn't suck, by Anna Burgess Yang. To learn more visit workbetter.media.","substackDomain":"https:\/\/www.workbetter.media","colorTheme":"blue","redirectUrl":""}}
Cloud storage services conveniently let you store and access documents, photos, videos, and more from any device. The best part? Many top providers offer free plans that are surprisingly capable.
But with so many options, how do you choose the right free cloud storage service? Let’s take a look at some of the leading free online storage services available today.
Google Drive
Google Drive offers 15 GB of storage, shared across Google Drive, Gmail, and Google Photos.
Thats one of the most generous free storage amounts, and the service offers seamless integration with Google Workspace (Docs, Sheets, Slides). Its also got a user-friendly interface and solid search capabilities.
The 15 GB is shared with other Google services, however, so heavy Gmail users or those backing up high-quality photos might find it fills up quickly. And Google being a large tech company, some users have privacy concerns regarding data handling.
Mega
At 20 GB, Mega provides the most generous permanent free storage tier among major providers.
Theres a strong focus on privacy and security with user-controlled end-to-end, zero-knowledge encryption on all plans, including the free one. The company cant see your files, in other words.
However, free accounts have transfer limits (you can download or upload a certain amount of data over a set period), and its unique encryption means if you lose your password or recovery key, your data is unrecoverable.
pCloud
pCloud gives you up to 10 GB of storage space, which is a decent amount of free space. It also features excellent built-in media players for streaming audio and video directly from the cloud. And it offers unique “Lifetime” paid plans, starting at a one-time $199 payment for 500 GB of storage.
However, to reach 10 GB on the free plan, youll need to perform certain actions like verifying your email address and installing mobile apps. Encryption and sharing options are also limited on the free tier.
Microsoft OneDrive
Although Microsoft OneDrive offers only 5 GB of storage on the free plan, its got excellent integration with Windows and Microsoft 365 (Word, Excel, PowerPoint).
Files can often be accessed directly from Windows File Explorer and theres a “Personal Vault” folder for extra security on sensitive files.
5 GB of free storage might not be enough for all your stuff, however, and there can be similar privacy considerations as Google Drive for some users.
Apple iCloud Drive
Like Microsoft, Apple iCloud Drive offers a meager 5 GB of storage. This storage is shared across iCloud Drive, device backups, Photos, and Mail.
That being said, its got unbeatable integration with Apple devices, making it essential for backing up iOS devices and syncing photos across the Apple ecosystem.
Those five gigabytes get consumed quickly by device backups and photos, though. Paid plans start at just a buck a month for 50 GB, making it kind of a no-brainer.
Dropbox
At a lowly 2 GB, Dropbox offers the least amount of free storage of the providers on our list here.
It is, however, a pioneer in cloud storage, known for its reliable and fast file syncing, simple user interface, excellent compatibility across platforms, and good third-party app integrations.
Sync.com
Sync.com offers 5 GB of free storage and boasts excellent security and privacy features, including end-to-end zero-knowledge encryption and compliance with features.
Theres also 30-day file versioning and recovery, no file size limits when using desktop or mobile apps, and you can earn more free space through referrals.
However, there are limited sharing capabilities on the free plan (number of links and daily downloads per link), and it lacks built-in document editing tools, instead relying on Office 365 integration.
Its inevitable that youll experience a slow period as a freelancer. No matter how long you’ve been at it, youll be faced with lulls from time to time where opportunities are coming in less frequently or not at all. Sometimes its not about you: Certain types of projects are impacted by seasonality, budget cycles, and economic conditions.
But other times, you could be the reason the engagements are starting to taper off, whether youve not staying up to date in your field or the competition is outpacing you.
To get out of these slumps, and prevent them from happening often, be intentional about what youre doing during these slow stretches.
Reframe your mindset
Ive experienced dips in work at numerous points throughout the 12 years Ive been freelancing. When it would happen initially, Id feel frustrated, embarrassed, nervous, and disengaged when a slow period started to happen beyond a week into a month. Through a mix of strategic maneuvering from me and new opportunities arising from long-term investments in marketing myself and building relationships, these periods didnt last long.
Its more than okay to experience all the feelings during these freelance slumps, but try not to spend too much time being distraught and instead take action based on whats in your control. As Barbara Corcoran famously said, The difference between successful people and others is how long they spend time feeling sorry for themselves.
When things begin to slow down, acknowledge what youre feeling but recognize its an opportunity to reset and take action.
Prioritize continuous progress
To move past these moments, I usually focus on making continuous progress on meaningful tasks for my business since I have more time to do soand it feels good. Im not alone in this: Professionals have reported more positive emotions on workdays where they made progress on meaningful work, and most often its small wins, according to a survey of 238 professionals by HBR authors Teresa M. Amabile and Steven J. Kramer.
For me, this might include small tasks like updating my website, reconnecting with a colleague, writing an article, taking a course, or whatever feels genuinely productive. Youre not just aiming to be busy but looking for work that helps generate momentum, improves upon your skill sets, attracts more projects, or simply provides creative fulfillment.
Ive found that this forward momentum helps me move past that stuck feeling and keeps me motivated to forge ahead. I recommend identifying up to five tasks youll work on everyday and aim for small ones that are possible to achieve in as little as a few hours or a couple of days. The hope is that the continued progress becomes ingrained as a lasting habit.
Expand your freelance offerings and income sources
In the midst of a dip, it might be time to reassess how youre reaching new customers and whether theres an opportunity to increase your visibility in a sustainable way. This could entail building a presence on another destination to reach your ideal like creating a Substack, being a guest on industry podcasts, or speaking at conferences.
Similarly, you might expand your platform as a freelancer to reach a different, complementary customer base with a service you may not have offered previously. During slow periods in my business, I try my best to treat it as a strategic opportunity to reflect and iterate, says, Gigi Robinson, a New York City-based creator, speaker, and author. Instead of spiraling or stressing about the lull, I dive into my analyticswhats working, whats underperforming, and how my audience is responding across platforms. I revisit my goals, experiment with new formats, and map out long-term ideas that dont rely on external brand campaigns or speaking invites.
For example, in 2023, she experienced a slowdown in inbound brand deals and speaking gigs. Rather than panic, I leaned into building something biggerand thats when I conceptualized Host of Influence, my new educational platform for creators, whether youre a newbie or a creator veteran, says Robinson. Its proof that sometimes the most impactful ideas are born when things get quiet. You just have to give yourself the space to listen and come up with a strategic game plan.
Chart a similar path for yourself by assessing where theres opportunity to market your services differently and provide new offerings that meet customer demands.
Network with other freelancers
Youre not the only professional to experience a period of stagnation. One of the most impactful things you can do at this point is reach out for help. Thats why it’s important that you use this time to connect with other freelancers to share how youre handling it and hear how theyve dealt with similar circumstances. Both parties can learn from each other about how to overcome these challenges, brainstorm new opportunities, and provide a fresh perspective on their career trajectory. You can find other professionals on freelancer marketplaces like MarketerHire and Behance, focused networking groups like ones for independent contractors in HR or engineering, and on Reddit, LinkedIn, and YouTube.
Whats worked for me is building relationships with colleagues that are ahead of me in their career or Id like to emulate aspects of what theyve accomplished professionally. Through their support, Ive been able to better anticipate some of the common challenges as a freelancer and get inspiration from how they’ve navigated their career over the years. The same can be said for fostering connections with new freelancers in their early days of self-employment as I have a lot of advice and support to offer them. Plus, these peers often give me exposure to a new viewpoint on the world of self-employment, helping me stay modern, adaptable, and motivated.
Pitch yourself
The loss of income is the scariest when youre in a work slump, and it could have a substantial impact on your day-to-day. One upside of not earning the income you expected is that it can be extremely motivating and drive you to put yourself out there in ways you hadnt anticipated.
“During a slow period, I reach out to past clients, prospects, collaborators, and peers with one question: Whats keeping you up at night right now? Their answers help me spot patterns and spark ideas I can quickly testwhether its a new workshop, course, or consulting offer, says, Hussain Bandukwala, a PMO and project management consultant, trainer, coach, and LinkedIn Learning Instructor. I also experiment with different outreach strategies, from lead magnets to paid ads, adds Bandukwala. Its not filler workits targeted R&D to evolve and optimize the business.
Dont become overly reliant on the right people and clients finding you, as attention youre generating for yourself professionally will ebb and flow due to a variety of factors. As an alternative, pitch yourself directly to existing clients who you already have a rapport with and to new organizationsand individuals as a chance to introduce yourself. Sometimes, the right nudge can make all the differenceyou never know who could benefit from your contributions, as long as your outreach is aligned to their needs.
