Screenshot of Apptronik’s Apollo humanoid robot handling basic home cleaning tasks in Google DeepMind’s robotics research lab.

John Koetsier

Putting a plant into a planter, placing snacks in containers, and sorting laundry successfully doesn’t necessarily mean you’re going to have a humanoid robot in your home next year. But the dream of offloading home cleaning, maintenance and maybe even cooking is getting a little more real as Google DeepMind showed off Apptronik’s Apollo robot obeying verbal commands and actioning tasks with objects it had never seen before.

In the video, Google shows robots opening Ziploc bags, inserting bread into the bag, sorting laundry by color, and manipulating odd-shaped real-world items that are sometimes squishy, sometimes difficult to pick up. The robots understand commands like “pick up the green block” or “sort this laundry into darks and whites,” and adapt to changes in the environment when their trainers move containers or objects they’re trying to pick up.

But they’re not fast.

“Sometimes they’re a bit clunky,” says Hannah Fry, a mathematician, broadcaster, and host of the DeepMind podcast. “But you have to remember that this idea of having a robot that can understand semantics, that can get a contextual view of the scene in front of it, that can reason through complex tasks is completely inconceivable just a few years ago.”

Google invested in Apptronik’s massive $403 million funding round earlier this year. In December last year, Apptronik announced a strategic partnership with Google DeepMind’s robotics lab to “bring together best-in-class artificial intelligence with cutting-edge hardware and embodied intelligence.”

To roughly over-generalize: Apptronik brings the robots, Google brings the brain. That brain has recently gotten a lot smarter with Gemini 3, and the robotics-specific version of Gemini, Gemini Robotics, is explicitly designed to support multiple embodiments – from dual-arm industrial robots to full humanoid form factors like Apollo – without retraining for each body.

The goal: a general purpose robot that can, essentially, do everything.

In other words, the Apptronik’s Apollo is being trained to do more than pick up boxes or repeat pre-programmed basic factory motions. It’s being taught to navigate the messy, unpredictable world that we humans inhabit: packing lunches, sorting laundry, opening unknown containers, and even responding gracefully when given previously unseen objects or tasks.

The latest hardware and software is getting much better at demonstrating that promise. Figure, which has shown its humanoid robot running smoothly and gracefully, has also shown Figure dealing with typical home challenges: putting dishes in the dishwasher, putting groceries away, and so on. The pace of development has massively accelerated in the past two years thanks to better AI, better hardware, and cheaper components.

If this most recent lab-based demo proves durable in the real world, it signals something fairly consequential. DeepMind and Apptronik are fusing high-quality humanoid hardware and foundation-model intelligence into a general-purpose robot that can perform a broad range of everyday physical tasks with minimal retraining. The result, potentially, could be the long-imagined “universal robot worker:” a cost-effective machine that can understand instructions, plan multi-step procedures, adapt to new objects and execute tasks with near-human dexterity.

Don’t hold your breath on that dexterity piece. Robots aren’t close yet: the Google demo in this video of putting a slice of bread into a Ziploc bag is all well and good … but you’ll note if you watch the video that the robot does not actually seal the Ziploc bag. That’s an astoundingly difficult thing to do that even humans struggle with sometimes.

Still, Apollo with Google’s DeepMind AI accomplishes at least four things:

1. Dexterity: delicately manipulating non-standard items, like a bag of chips
2. Generalization: handling objects that it had never seen before correctly
3. Natural-language control: obeying verbal commands that require a significant world model, like “put the green block in the orange tray”
4. Long-horizon planning: planning out multiple steps to accomplish a task

But there’s a long way to go. Humanoid robots will need to be faster in accomplishing tasks: right now they all look like they are moving in slow motion when handling objects and doing work. That means better hardware: joints, muscles (actuators), and control systems.

In addition, they need better training methods.

“These robots take a lot of data to learn these tasks,” says Kanishka Rao, director of robotics at Google DeepMind. “So we need a breakthrough where they can learn more efficiently with data.”

That’s interaction data and manipulation data: data that robotic brains can take and use to learn how to do tasks that they’ve never been faced with before.

And finally, of course, they need to be guaranteed safe to use in human environments, where they might encounter people – including children – and pets.

Plus, of course, grandma’s good china.

Shoppers want to buy secondhand; the challenge is making it effortless.

getty

Secondhand fashion has moved from niche hobby to mainstream shopping behavior. What was once dominated by thrift stores and consignment racks has expanded into a sprawling digital ecosystem with marketplaces, social commerce and peer-to-peer platforms. Consumers increasingly look to resale for affordability, sustainability and personal expression. Yet even as enthusiasm grows, the market’s full potential remains constrained by a structural weakness: discovery.

Shoppers may like the idea of secondhand. They may even begin with good intentions. But the moment they try to find a specific item, many run into the same obstacles — fragmented platforms, inconsistent listings, unpredictable sizing and the need to manually piece together information that traditional e-commerce delivers effortlessly.

To understand what it will take for resale to scale, it’s important to understand why discovery is so difficult in this category and how some innovators are attempting to solve it.

The Discovery Dilemma at the Heart of Resale

The secondhand market offers clear benefits. In an inflationary climate, it gives shoppers access to higher-quality garments at more approachable prices. It also resonates with consumers who want to reduce waste without fundamentally changing their style habits. But interest alone does not translate into consistent behavior.

Unlike traditional retail, secondhand doesn’t operate on a clean, structured product catalog. Two identical items may be photographed from different angles, described with unrelated keywords or tagged inconsistently. Lighting varies from professional studio setups to bedroom mirror selfies. A dress might be listed as “midi,” “knee-length,” or “summer dress,” while a jacket might be described only as “cute coat.” These variations make precise search nearly impossible.

Fragmentation deepens the complexity. The item someone wants might be on Poshmark, or eBay, or ThredUp or an entirely different niche marketplace. Without a unified way to search across platforms, shoppers either make peace with incomplete results or abandon the effort altogether.

For many consumers, the value proposition is clear — but the experience is not yet convenient enough to compete with the efficiency of buying new.

Why Technical Innovation Matters Now

As the category matures, a number of companies are trying to simplify what makes resale so hard. Visual search is one promising tool, particularly as more fashion inspiration starts with images rather than text. But applying visual search to secondhand is significantly more complicated than applying it to new retail.

This is the context in which Beni Lens, a new visual search tool from the resale-focused startup Beni, has emerged. The company argues that secondhand requires its own technical approach, given the variability of real-world images and listing data. Co-founder and CEO Kate Sanner explains the challenge: “There’s no clean product catalog to draw from. So instead of matching to standardized images, our model has to interpret the intention behind an image and express that in attributes that can be mapped to millions of messy and varied listings.”

Beni built an ingestion engine to maintain a real-time catalog of hundreds of millions of listings and a search system that translates a single photo into structured attributes like silhouette, color, fabric and design details. From there, it identifies similar items across marketplaces and organizes them into a navigable feed.

What matters for the broader market is not only the specifics of how Beni Lens works, but what its existence represents: a shift toward tools built expressly for secondhand’s complexities, rather than adaptations of systems designed for traditional retail.

Where Resale Still Falls Short for Mainstream Shoppers

Despite growing interest, many mainstream shoppers still hesitate to explore resale regularly. The intent is there, but the process feels unpredictable. Searching requires patience; sizing requires guesswork; comparing prices across platforms requires persistence. Even experienced thrift shoppers acknowledge that the process can feel like a scavenger hunt.

Sanner summarizes the issue bluntly: “The biggest barrier to choosing secondhand isn’t desire — it’s effort.”

Until that effort decreases, the resale market will continue to rely on a self-selected group of enthusiasts who enjoy the chase. To reach the next stage of growth, the experience will need to support shoppers who want outcomes, not adventures.

The Long Tail Advantage — and Why It’s Underutilized

One of the greatest strengths of the secondhand market is its access to the long tail of fashion: discontinued styles, archival pieces, vintage finds, rare colorways and items that never made it to mass retail distribution. No traditional retailer can compete with this breadth.

But this long tail only unlocks value when consumers can actually find what they want — or discover what they didn’t know they were seeking. Without tools that make this inventory accessible, much of it remains invisible.

Better discovery doesn’t just improve the user experience. It increases sell-through rates, keeps garments in circulation longer and expands the total market of shoppers willing to consider secondhand as a first choice rather than a fallback option.

What the Next Phase of Resale Could Look Like

The past decade of resale was defined by the rise of major marketplaces. The decade ahead is likely to be defined by infrastructure — better search, cleaner data, stronger fit tools, smarter personalization and tighter integration with the moments when style inspiration actually happens.

Sanner imagines a future where resale is embedded directly into everyday shopping moments. “One click away whenever and wherever inspiration strikes.”

The specifics will vary by company, but the underlying idea is shared by many technologists in the space: a world where secondhand is as intuitive as any other form of e-commerce.

Secondhand doesn’t face a demand challenge. It faces a usability challenge. The companies that solve discovery — whether through visual search, meta-search, improved data standardization or new forms of personalization — will accelerate not just market growth, but a broader cultural shift in how people shop.

Australia banned some major social media apps from serving kids under the age of 16.

Getty Images

Australia may have banned social media apps for under-16s, but kids are finding ways around the ban already. Will the government now play Whack-a-Mole with new and smaller social apps?

Australia recently banned social media for kids under the age of 16, so Facebook, Instagram, TikTok, Snapchat, X, YouTube, Reddit, Twitch, and Kick are are all off limits for millions of Australian teens. This is the first major social media ban for kids on the planet. But kids that want their social fix – or their short video hit for the day – are finding alternatives. And it’s not even hard.

According to Apptopia, an app analytics company, here are the top 10 apps by downloads in Australia, on all platforms and in all categories for yesterday:

1. Lemon8 – Lifestyle Community
2. Yope: friends-only pics
3. Australia Post
4. WhatsApp Messenger
5. ChatGPT
6. Meta Horizon
7. Coverstar – Positive Social
8. Shop: All your favorite brands
9. Temu: Shop Like a Billionaire
10. myGov

Five of the top 10 are clearly getting a massive positive boost from the social media ban. The top app, Lemon8, is literally made by the company behind TikTok, a banned app. Lemon8 offers “a lifestyle community focused app powered by TikTok, where you can discover and share authentic content on a variety of topics such as beauty, fashion, travel, food, and more.”

It allows photo editing and sharing, just like Instagram – another banned app – and sounds pretty social.

WhatsApp Messenger is up in downloads as well, and it offers messaging, calling, groups, and video chat. The biggest growth, however, comes from Meta Horizon, which is social gaming app that promises users they can “step into Metaverse where you can play, explore and connect with friends in a variety of community created worlds.”

That also sounds pretty social.

Yope, a “friends-only” photos app, is another Instagram competitor. Coverstar is a social video sharing app, much like TikTok. Again, pretty social.

The upshot is clear: while Australia has banned the big social media networks that get all the press, teens are finding new platforms to replace them. And while Australia promises severe monetary penalties to the big social platforms if they allow teens on them, those big platforms have the resources to be able to monitor and build systems to try to exclude under-16 kids. The smaller platforms wouldn’t have the same resources, if Australia decides to extend the ban to all social apps.

Other big social platforms that are not currently banned include:

• Discord
• Steam and Steam Chat
• Roblox
• Pinterest

Australia is an interesting test case that other countries will be watching closely to see if this in the best interests of children, and whether it makes kids safer. I’ve definitely seen many in favor in other countries.

“Global platforms have become the new, unchosen parents of our children—shaping their identity, their worldview, their values,” says Jamaican technologist Chukwuemeka Cameron on LinkedIn. “All while extracting their data for profit.”