Ami Dror

𝗪𝗵𝗮𝘁 𝗺𝘂𝘀𝗰𝗹𝗲𝘀 𝘄𝗲𝗿𝗲 𝘁𝗼 𝗳𝗶𝘁𝗻𝗲𝘀𝘀 𝗶𝗻 𝘁𝗵𝗲 𝟮𝟬𝟬𝟬𝘀. 𝗧𝗵𝗲 𝗯𝗿𝗮𝗶𝗻 𝘄𝗶𝗹𝗹 𝗯𝗲 𝘁𝗼 𝘄𝗲𝗹𝗹𝗻𝗲𝘀𝘀 𝗶𝗻 𝘁𝗵𝗲 𝟮𝟬𝟯𝟬𝘀. Last week Atlas came out of stealth with $14M in funding to launch a multi-modal brain-sensing wearable next year. The behind-the-ear nanosensor tracks mental clarity in real time, decoding how your behaviors, from training to phone use, affect focus and cognitive performance. Built by neuroscientists from Oxford and Cambridge, the founding team includes leaders from ŌURA and Kernel. We’ve spent years tracking everything below the neck, sleep, recovery, nutrition. Now it feels like the focus is moving above it, towards brain prevention. → 55M people worldwide live with dementia, with nearly 10M new cases each year. → Global dementia costs exceed $1.3T annually and are set to double by 2030. → In the U.S., Alzheimer’s and other dementias will cost nearly $1T by 2050. → Up to 40% of cases could be prevented by tackling risk factors like inactivity, hypertension, and obesity. A Nature Medicine study showed that a simple blood test measuring plasma p-tau217 can detect Alzheimer’s 15 to 20 years before symptoms with 96% accuracy. That changes how we think about prevention, catching decline before it starts. And it’s showing up in culture and business too: → The brain health market is set to jump from $10B in 2024 to $32B by 2034. → Retailers like Planet Organic, Boots UK, and Erewhon now have braincare sections. → Cognitive nutrition and nootropic drinks are some of the fastest-growing wellness categories. Atlas is joining a fast-growing wave of companies focused on cognitive health: 🧠 Muse® by Interaxon Inc. has raised $60M for its Athena headset using EEG and fNIRS to measure brain oxygenation and cognitive load, basically VO2 max for the mind. 💤 Pulsetto raised €2M for a vagus-nerve wearable that helps manage stress, focus, and sleep. 🔬 Flow Neuroscience backed by $13M, built a CE-approved tDCS headset that treats depression while boosting cognitive performance. 💡 Elemind Technologies, Inc. raised $12M for a wearable that uses personalized brainwave stimulation to improve sleep, focus, and relaxation 💤 FLOWTIME backed by $5M, builds EEG meditation headsets that visualize brainwaves and guide users into deeper focus and calm. 🎮 MindMaze a Swiss unicorn valued above $1.5B, blends neuroscience and VR to create neurorehabilitation and cognitive training tools used in hospitals and homes. Once you’ve optimized your physical health, the next question is how to protect the system that drives it all. Extending cognitive health by just five years could save global economies trillions in healthcare and lost productivity. Sharper brains mean longer productive years, more creativity, and sustained performance. From muscle to mind, from performance to prevention, brain prevention is the natural next step in proactive health, where longevity and productivity meet.

315

78 Comments

Some big news after yesterday's announcement (let's hope Miliband realises we need cheap and abundant energy). NVIDIA Announces £2 Billion Investment in the United Kingdom AI Startup Ecosystem New Capital Supports U.K. AI Infrastructure and Startups to Spark New Industries · NVIDIA commits £2 billion to catalyze the U.K.’s AI startup ecosystem and accelerate the creation of new companies, jobs and globally transformative AI businesses. · Accel, Air Street Capital, Balderton, Hoxton Ventures and Phoenix Court will join NVIDIA’s investment in the U.K. market, underscoring their commitment to foster the U.K. AI ecosystem. Scaling AI companies in the U.K. has been challenging due to limited access to supercomputing, constrained venture capital outside London, rising energy costs and difficulty for VCs to access leading academic institutions, where many researchers are also entrepreneurs. NVIDIA, in collaboration with Accel, Air Street Capital, Balderton Capital, Hoxton Ventures and Phoenix Court, will accelerate the U.K. AI ecosystem by providing new capital for AI startups. “This is the age of AI — the big bang of a new industrial revolution,” said Jensen Huang, founder and CEO of NVIDIA. “The United Kingdom is in a Goldilocks moment, where world-class universities, bold startups, leading researchers and cutting-edge supercomputing converge. There has never been a better time to invest in the U.K. — AI is unlocking new science and sparking entirely new industries. With new capital and advanced infrastructure, we are doubling down to empower the U.K. to lead the next wave of AI innovation.” “NVIDIA’s investment is a major vote of confidence in the U.K. both today and long into the future,” said U.K. Prime Minister Sir Keir Starmer. “By backing our startups, empowering our researchers and connecting capital with talent, this partnership will create jobs, spark new industries and ensure the U.K. remains at the forefront of global AI leadership.” NVIDIA is committed to working closely with the U.K.’s leading company founders and VCs — especially in London, home to world-renowned academic AI labs — and partnering with top U.K.-based funds to ensure capital and compute reach the most ambitious entrepreneurs and most dynamic regions, including the U.K. government’s newly announced AI growth zone. Following NVIDIA’s recent commitment to manufacturing up to a half-trillion dollars’ worth of AI supercomputers in America, the investment will be domiciled in the United States and activated in the United Kingdom, celebrating the transatlantic partnership between the two countries. NVIDIA is purposefully investing in the U.K.’s future, unlocking opportunities for the next generation of British innovators. “We are in the midst of a seismic technology shift as people and companies around the world increasingly depend on more intelligent hardware and software,” said James Wise, partner at Balderton Capital.

44

8 Comments

AI voice agents are finally starting to feel more human—but they still break down in noisy environments. Today we launched VIVA, Krisp’s voice isolation SDK built specifically to solve this problem. Already processing over 1𝐁𝐈𝐋𝐋𝐈𝐎𝐍 minutes a month, VIVA improves turn-taking and speech detection by filtering out background noise and voices before the audio reaches an AI agent. It’s already helping systems reduce dropped calls by 50% and increase CSAT by 30%. Thanks to Kyt Dotson and SiliconANGLE & theCUBE for the write-up and conversation: https://lnkd.in/emD8-uKF

55

6 Comments

Ramblr.ai now supports Meta Wearables SDK ! Ramblr.ai’s Data Engine now supports the Meta Wearables Device Access Toolkit for Meta Ray-Ban smart glasses—unlocking camera and audio access for real-world #AI experiences. This marks a mainstream moment for #Physical AI: - Capture video from smart glasses  - Train Skills on real-world processes with Ramblr.ai Data Engine - Deploy AI Agents that deliver information when and where it’s needed Real-world sensors demand real-world understanding—and that’s exactly what Ramblr provides. Developers can now build AI that’s grounded in reality, not just data centers. Learn more: https://ramblr.ai #PhysicalAI #AI #Meta #SmartGlasses #RamblrAI #Wearables #EdgeAI #Ramblr.ai

160

3 Comments

Many (including me) discovered #Ghibli through modern reinterpretations. Hayao Miyazaki, co-founder of Studio Ghibli, called AI an "insult to life itself." But more than that, it is melting GPUs, machines that run AI! Since Ghibli-style art has taken over, we at Unstop couldn't help but join the trend. It’s undoubtedly one of the cutest trends ever, but we can’t overlook the different angles of controversy it brings. OpenAI’s CEO Sam Altman said, "Our GPUs are melting." Why? Millions are generating AI images, pushing GPUs to their limits. - Power Surge: Training a single AI model can emit as much CO₂ as five cars in their lifetime. - Water Crisis: Every AI-generated image can use 5 to 50 liters of water—that’s your entire daily drinking supply gone in seconds. - Overheating GPUs: OpenAI’s servers are running at full throttle, causing extreme GPU strain. More AI generation= more electricity, more heat, and higher cooling costs. - Carbon Footprint: Every AI prompt adds up. The computing power behind ChatGPT-4 consumes 10x more energy than its predecessor. As someone who builds for creators and learners at Unstop, I see AI as both a game-changer and a challenge. It opens doors but also raises tough questions. AI is a double-edged sword powerful, but ethically murky. We’ve known this for a long time, but the line between creativity and automation is getting thinner. Where do we go from here? #sustainability #AI #Ghibliimage

112

3 Comments

✨ Highlight from my panel at the Vapi #Vapicon! ✨ For the first time, we’re seeing the rise of highly specialized models, not just general-purpose ones. 🎙️ Imagine speech generation models tuned for influencer accents. 🎥 Or multimodal narration built specifically for video generation. 📱 And consider how differently people speak on TikTok versus on a customer support call. This context matters, and models are starting to reflect that. What excites me most is that we’re moving from “What does this model score on benchmarks?” to “What can I actually do with this model?”

75

1 Comment