What would it take to run powerful AI models like GPT-4 continuously and with low latency on edge devices like your smartphone, earbuds, or even smart glasses? @semron_ai pioneers analog AI chips based on memcapacitors, encoding the weights and biases of machine learning models by hardware design to reach unparalleled energy efficiency. Since our previous interview in the summer of 2022, SEMRON went through the @IntelIgnite startup program and raised a €7.3 million seed round by @join_capital, @SquareOne_vc, @otb_ventures, @OnsightVentures, Hermann Hauser, and others. I had the pleasure of speaking again with the co-founder and CEO, @Aron_Kirschen, about what it takes to develop analog AI chips, from the foundry process to a demo board, and how Intel Ignite accelerated their journey: future-of-computing.com/aron…

Last year, @INAMBerlin organized the first Future of Computing conference in Berlin I had the pleasure of giving a talk about investment opportunities in the AI Hardware Stack – it's now online on YouTube 🤗 TL;DR: We're seeing the unbundling of semiconductors, going from CPUs and GPUs to various kinds of novel chips specialized for AI applications There is potential for innovation at every level of the stack – here are a couple of examples: 1) Novel lithography methods – Lace Lithography is developing breakthrough chip patterning technologies 2) Analog AI chips – @semron_ai is developing ultra-efficient chips for AI inference using memcapacitors 3) Faster and larger on-chip – RAAAM Memory Technologies is developing high-density, cost-effective SRAM 4) Optical interconnects for chiplets – NcodiN is developing nanolasers for ultra-fast and energy-efficient die-to-die interconnects 5) Optical computing – @Akhetonics is developing the first all-optical general-purpose processor youtu.be/lXtgDTLDlIE?si=vaOp…