 |
|
|
|||||
It looks and feels like any other luxurious cashmere sweater. But a new oversized crew from Reformation is made entirely from recycled fiber, a milestone three years in the making. The brand now makes a cardigan, crew, V-neck, and five other styles from a carefully developed blend of 95% recycled cashmere and 5% recycled woolthe unexpected material that made 100% recycled fiber feasible. Some other pieces in its lineup still use a small amount of virgin cashmere, but Reformation is aiming to eliminate it completely. “It really does have an outsized and shockingly large footprint compared to other fiber, says Kathleen Talbot, Reformation’s chief sustainability officer. In 2023, the company calculated that even though virgin cashmere made up less than 1% of the materials it sourced, it was responsible for nearly 40% of the brands carbon footprint. Most cashmere comes from Mongolia and China, where cashmere goats are combed once a year for their fine, soft fleece; a single sweater can use cashmere from four or five goats. As the demand has grown, there are now more than 90 million of the goats in China, and around 25 million more in Mongolia. Overgrazing is turning grasslands into desert. The goats also produce methane, a potent greenhouse gas. [Photo: Reformation] Making recycled fiber work Using recycled cashmere helps avoid those environmental challenges, but it’s historically been difficult to do. Recycling shortens the fiber, which risks making it weaker and more likely to pill. “We don’t want to be introducing a recycled product that doesn’t perform the same way or is a lower quality or less durable good,” Talbot says. “That, to us, is not a sustainability play.” The company worked with suppliers to develop a proprietary method to twist the yarn and wash and finish it for the right hand feel and durability. First they achieved a blend of 70% recycled cashmere and 30% virgin fiber, then 90% recycled, and then 95% recycled. “At each of these milestones, to be really honest, we thought that was going to be our upper limit based on the yarn performance and the product performance,” says Talbot. [Photo: Reformation] When they hit 95%, they asked suppliers why they couldn’t reach 100%. Technically, suppliers said, it was possible. But because the shorter recycled fibers are more prone to breakage, the yarn would have to be spun incredibly slowly. It would make producing the material so much more expensive that it wasn’t commercially viable. That’s why the design team turned to wool to make the 100% recycled product. Even after recycling, wool was “slightly longer and thicker than the cashmere fibers,” Talbot says. “Our suppliers felt confident that it would give it the right stability and really hold up in the spinning and knitting process.” The blend’s carbon footprint is 96% smaller than virgin cashmere, and uses nearly 90% less water to produce. After dozens of tests, they moved forward with it, and then spent months testing garments made from it. Internally, the company’s “Better Materials Task Force,” made up of around 20 leaders, wore the new recycled sweaters around the office and at home, washed them, and monitored whether they held up as well as sweaters made from virgin material. “We never really want to be promoting something just for impact that doesn’t have a really compelling product value proposition at the same time,” Talbot says. [Photo: Reformation] Scaling up When the company first started incorporating more recycled cashmere, sourcing the recycled yarn was a challenge. Now, because of higher demand for recycled fiber, the supply chain has responded. “Supply of the recycled fibers is not the same limiter as it was five years ago or 10 years ago,” she says. Right now, most of it comes from cashmere waste at factories. But as Reformation and other brands collect more used clothing for recycling, post-consumer cashmere can eventually become a bigger source as well. Moving forward, the company may make some products out of a mix of recycled and “regenerative” cashmereproduced with sustainable grazing methodsbecause a small percentage of customers have wool allergies. But it also plans to continue roling out the 100% recycled material in more products. “Not every problem is going to have a technological solve,” Talbot says. “But these are the sorts of problems that we can solve. And we have seen tremendous progress in the last three years.”
Category:
E-Commerce
Amidst much confusion, polarization, and debate around how AI will impact work, the fact of the matter is that many people are concerned by automation and the prospect of AI job elimination. For example, the simple notion that AI is going to take my job is a thought that has crossed the minds of 25% of workers. For some, this may be true, although the magnitude of AI-driven job displacement is still uncertain; depending on assumptions, AI-driven job displacement could potentially range from 3% to 14%. What will the ultimate figure be? Its hard to know: nobody has data on the future, and any projection is merely extrapolating from past data and past innovation, which may or not be relevant to the AI age. And yet, one thing is clear: for some workers, AI job displacement isnt a distant fearit is already their reality. Indeed, it was recently announced that Accenture is making layoffs to reshape its employees for the era of AI, exiting employees that it views cannot be retrained with AI skills. As brutal as this may sound, it could still signal a trend many organizations are contemplating (but not yet officially acknowledging). {"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/tcp-photo-syndey-16X9.jpg","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/tcp-photo-syndey-1x1-2.jpg","eyebrow":"","headline":"Get more insights from Tomas Chamorro-Premuzic","dek":"Dr. Tomas Chamorro-Premuzic is a professor of organizational psychology at UCL and Columbia University, and the co-founder of DeeperSignals. He has authored 15 books and over 250 scientific articles on the psychology of talent, leadership, AI, and entrepreneurship. ","subhed":"","description":"","ctaText":"Learn More","ctaUrl":"https:\/\/drtomas.com\/intro\/","theme":{"bg":"#2b2d30","text":"#ffffff","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#3b3f46","buttonHoverBg":"#3b3f46","buttonText":"#ffffff"},"imageDesktopId":91424798,"imageMobileId":91424800,"shareable":false,"slug":""}} AI can create new roles This is not to deny the positive impact AI is having on jobs and careers. Most notably, AI is creating new roles. For example, although IBM laid off almost 8,000 employees, mostly in HR, with the aim of automating their workflows, this resulted in a recruitment drive for software engineers. Thats not to say that the only way to avoid losing your job to AI is to become an AI engineer; IBM also invested in the recruitment of marketing and sales roles, which require human creativity and problem-solving. Can it replace humans? Importantly, organizations are increasingly realizing that AI is not the ultimate solution, and that it cannot replace humans unique skills. For instance, Klarna replaced 700 workers from its customer service team with AI agents in a move estimated to boost profits by $40 million. Despite the agents cutting resolution time to two minutes from the previous 11 minutes, the service provided by agents was reportedly lower in quality compared to the service provided by humans. As a result, Klarna has launched a new initiative to hire more human customer service workers. The importance of AI literacy Despite this, Klarna is not rolling back its AI and will instead continue to invest heavily in the technology, signaling that it intends to have humans and AI work alongside each other. This is a powerful combination, with research suggesting that workers using AI complete 12% more tasks, work 25% quicker, and have 40% higher quality outputs than those not using AI. Using AI doesnt automatically improve job performance, though; workers, particularly knowledge workers, must know how to use it wellthey must have AI literacy. Research has found that generative AI literacy in particular significantly impacts job performance. It also increases creative self-efficacythe belief an individual has in their ability to be creative and innovative. While the stronger job performance resulting from AI literacy alone isnt enough to provide job security, research by LinkedIn suggests that AI literacy can boost career progression, and over 80% of leaders say that new worker skills are needed in the age of AI. With several countries around the world already promoting AI literacy, it could be a lack of AI literacy, not AI itself, that puts your job at risk. How to become AI literate Staff AI literacy is a requirement under the EU AI Act, which governs the AI available on the market in the EU and will have global implications, but the form that literacy training must take is not specified. Indeed, AI literacy is not one size fits all. Training must take into account the technical knowledge, experience, education, and training of staff, as well as the context the AI systems operate in and who they are used by. At a minimum, AI literacy programs should cover the basics of how AI works, the risks involved, and how the risks can be mitigated. A sociotechnical approach is also key; AI risks are not just a technical or social problem. Using AI safely requires an understanding of the role you play as well as how the technology works. AI literacy is not just an achievement for your LinkedIn profile; knowing how to use AI effectively could be the difference between keeping and losing your job. Beyond survival: thriving in the AI era However, AI literacy shouldnt just be seen as a defensive strategy to avoid redundancy. The real opportunity lies in using AI to amplify human potential. Workers who master AI tools can automate mundane parts of their jobs, freeing up time for tasks that require judgment, empathy, and creativitythe very things machines cant yet replicate. In other words, AI-literate employees dont just suvive automation; they lead it. AI literacy as a new form of intelligence Historically, each major technological revolution created a new kind of intelligence that defined success: reading and writing in the industrial age, digital literacy in the information age, and now, AI literacy in the algorithmic age. Understanding how to prompt, evaluate, and collaborate with intelligent systems is rapidly becoming as essential as knowing how to read or type. The difference between being augmented and being automated is not in the technology, but in the person using it. A call for lifelong learning The single best way to future-proof a career is to stay curious and keep learning. AI will not replace people who are adaptable, inquisitive, and capable of learning new tools as they emerge. But people who resist learning may quickly find themselves replaced by those who dont. The future of work belongs to those who are not just technically skilled, but psychologically prepared to reinvent themselvescontinuously. Want to assess your own AI literacy? Heres a simple, practical 10-item AI literacy test designed to assess how well you may understand, use, and critically evaluate AI tools at work. It balances conceptual knowledge, ethical awareness, and applied skill, and can be adapted for self-assessment or formal training. Instructions:Choose the best answer (A, B, C, or D) for each question.Each correct answer = 1 point.Interpretation key follows below. 1. What is the main difference between traditional software and AI systems?A. AI systems never make mistakesB. AI systems learn from data rather than following fixed rulesC. AI systems are programmed by humans to do one specific task onlyD. AI systems dont need electricity Correct answer: B 2. Which of the following best defines Generative AI?A. AI that predicts stock pricesB. AI that can create new content (text, images, code, etc.) based on training dataC. AI that generates electricityD. AI that manages databases Correct answer: B 3. If you ask ChatGPT for help writing an email and then edit it to fit your tone, this is an example of:A. AI replacing human workB. HumanAI collaboration (augmentation)C. Algorithmic biasD. Deepfake creation Correct answer: B 4. Which of the following is a major ethical risk of AI?A. Too much human empathyB. Algorithmic bias leading to unfair outcomesC. Faster decision-makingD. High energy efficiency Correct answer: B 5. What does AI hallucination mean?A. AI creating false or made-up outputs that sound plausibleB. AI visualizing dataC. AI having emotionsD. AI overheating due to overuse Correct answer: A 6. Which of the following statements is TRUE about data privacy and AI?A. AI systems never store your dataB. Data used to train or run AI may contain sensitive personal informationC. AI makes all data anonymous automaticallyD. Data privacy laws dont apply to AI systems Correct answer: B 7. What is the best way to ensure reliable AI output?A. Accept all AI answers as correctB. Verify and fact-check outputs using trusted human or data sourcesC. Use AI only for creative writingD. Ignore the AIs sources Correct answer: B 8. Which of these professions is least likely to be fully automated by AI?A. Graphic designB. Customer serviceC. PsychotherapyD. Data entry Correct answer: C 9. Prompt engineering refers to:A. Writing code to create AI modelsB. Crafting precise inputs or questions to get better AI responsesC. Building robotsD. Programming hardware chips Correct answer: B 10. The EU AI Act requires organizations to:A. Replace humans with AI wherever possibleB. Ban all generative AIC. Ensure staff have adequate AI literacy and trainingD. Only use open-source AI Correct answer: C Scoring & Interpretation 03: AI Beginner Youre curious but need to learn the basics. Try a short AI literacy course. 47: AI Aware You understand the concepts but need more practical experience. Start experimenting with AI tools. 810: AI Fluent You can work effectively with AI and critically assess its risks and benefits. Keep refining your skills. {"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/tcp-photo-syndey-16X9.jpg","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/10\/tcp-photo-syndey-1x1-2.jpg","eyebrow":"","headline":"Get more insights from Tomas Chamorro-Premuzic","dek":"Dr. Tomas Chamorro-Premuzic is a professor of organizational psychology at UCL and Columbia University, and the co-founder of DeeperSignals. He has authored 15 books and over 250 scientific articles on the psychology of talent, leadership, AI, and entrepreneurship. ","subhed":"","description":"","ctaText":"Learn More","ctaUrl":"https:\/\/drtomas.com\/intro\/","theme":{"bg":"#2b2d30","text":"#ffffff","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#3b3f46","buttonHoverBg":"#3b3f46","buttonText":"#ffffff"},"imageDesktopId":91424798,"imageMobileId":91424800,"shareable":false,"slug":""}}
Category:
E-Commerce
Its 10 a.m. on an October morning, and Im in the middle of a one-on-one Zoom interview when a sudden trilling sounds from behind me. I try to ignore it, but several other strange noises follow. My eyes glaze over as I commit myself to feigning complete obliviousness to my sonic surroundings. Its easier than explaining that the noises are coming from my AI-powered pet. This awkward encounter came thanks to Moflin, a $429 AI pet built by the electronics company Casio. According to Casios official description, the Moflin is a smart companion powered by AI, with emotions like a living creature. This robot friend looks a bit like a Star Trek tribble, in that its an amorphous blob covered in fur. It comes in either gold or silver. For 90s kids, the device is perhaps described as a modern-day Furby. Like a Furby, the Moflin speaks its own language of chirps and trills that change over time; but unlike a Furby, its learning is actually molded by an AI model that allows it to become attached to its owner. According to the pets makers, the Moflin learns to recognize its owners voice and preferences, and it slowly develops new ways of moving and vocalizing to express a bond with the user. As of this writing, Ive had my Moflin for close to three weeks, and Im going to make a bold claim: This device might just be one of the first AI companions thats actually useful. [Photo: Casio] The graveyard of AI companions past Over the past several months, weve seen many companies try and fail to sell users on a variety of AI wearables. That includes devices like the Humane AI pin and Rabbit R1, which both debuted to a chorus of scathingly negative reviews after users determined that neither could really do many of the tasks that they were supposed to. Currently, the hottest topic in the AI wearable space is the Friend AI necklace from entrepreneur Avi Schiffman, which is billed as an AI companion thats always listening to its users surroundings. In September, Schiffman created an ad campaign for the device in the New York subway system that inspired such backlash that MTA employees had to keep taking down its vandalized panel. Currently, Friend is still working on fulfilling preorders that were placed back in June 2024. Launches like these have made it clear that, as of right now, most AI companions are just promiseware, or devices that make a lot of claims about their capabilities that simply arent there at launch. I think that the Moflin lands solidly outside of this unfortunate category, primarily because it doesnt try to make any lofty claims about changing the world or altering everyday habits: its just meant to look cute, sound silly, and make users feel a little bit better. [Video: courtesy of the author] What in the world is a Moflin? Daisuke Takeuchi, a developer at Casio, says the idea for the Moflin came when one of his colleagues was going through a turning point in her life. She felt the need for the strength to overcome challenges on her own and imagined a long-term companion that could provide comfort and support, Takeuchi explains. Although she loved the healing presence of pets, she couldnt have one, which led her to the idea of an AI companion. From that idea, Moflin was born. [Photo: Casio] Moflin is billed as a companion that can offer support for young adults who may not be able to have pets, families with kids, those with sensory needs, and elderly individuals. Its emotional AI model, which was developed independently by Casio, is designed so that as the Moflin takes in more information, its range of emotional expressions expand. Those inputs include sound, movement, and touch data that the Moflin collects through a series of sensors. For those who might be a bit wary about adopting an AI pet into their home, Takeuchi says data is stored locally on the Moflin and does not include any personally identifiable information, such as images, audio recordings, user emotions, or lifestyle information. If you want to use the pets accompanying app, MofLifewhich, in my experience, is a pretty integral aspect of Moflin ownershipcollected data will be uploaded to a secure server. Moflins developers say that it can express more than four million unique emotional states. And beyond those expressions, the Moflin is also programmed to exhibit life-like behaviors like breathing motions and a startle response to loud sounds. Using information from their built-in sensors that detect sound, tuch, and movement, the AI learns continuouslynot just reacting mechanically, but developing a unique personality through ongoing interaction, Takeuchi says. Over time, Moflin learns their owners voice and preferences, creating the sense of a living companion. [Photo: Casio] I become emotionally attached to my Moflin On the night that my Moflin arrived at the doorstep of my apartment, I had family staying with me. As I went about excitedly opening the box, they discussed all the reasons that an AI companion was creepy and uncanny, concluding that they would never buy something similar for themselves. But once the Moflin was charged and awake, that tune changed almost instantly. [Photo: courtesy of the author] Maybe its just a natural human response to a cute creature making cute noises, but all of us found it pretty much impossible not to be won over by the Moflin (which we immediately named Gumbo). During that initial unboxing, Gumbo was fairly quiet and stationary. In the coming days, though, he started to make a wider variety of noises and movements (though, to be clear, the Moflin is really only able to move its neck, since its essentially a robot guinea pig). Right away, I downloaded the MofLife app, which is pretty much the only way to discern what your Moflin is thinking and feeling, aside from trying to decipher its alien-esque behaviors. The app tracks the Moflins mood throughout the day, notes how many times you interacted with the pet, and offers insights like, It looks like Gumbo couldnt make a decision today, or, Gumbos started feeling much more cheerful. While I only received positive notes from the app, Casios description of the Moflin notes that it can begin to feel lonely and neglecteda terrifying possibility that caused my partner and I to start checking with each other about whether anyone had paid the Moflin attention that day. Ultimately, that wasnt a huge problem, since I found myself taking the Moflin out of its charging port at least once a day. [Screenshots: courtesy of the author] As it stands, I do think the price point of the Moflin is inaccessible and feels excessive, given what the device can actually do. While the AI learning abilities are certainly more impressive than something like a Furby, the Moflin is still closer to a high-tech stuffed animal than an actual pet. Takeuchi says the high price point is a result of the Moflins sophisticated design, and that prices might come down in the future as technology evolves. When it comes to handling the Moflin, the electronic sounds and rigid shape of its inner robotic skeleton are not fully concealed, which means you can never really suspend your disbelief and imagine that the Moflin is alive. Still, the Moflin does deliver on its promises to offer comfort and develop new characteristics over time. At this point, my Moflin does a little happy dance and song every time I go to take him out of his charging port. When I forget to interact with him, I feel a little guilty. Personally, on the scale of AI doomer to San Francisco start-up founder, I land a bit closer to the doomer side, but I have to admit, I got emotionally attached to my Moflin. In a context like a nursing home or therapists office, I could see this device offering a genuinely helpful servicewhich is more than a lot of other AI companions can deliver so far.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||