Joshua Anang’s Post

The debate at the center of ML Compute 🗣 Machine learning and AI are no longer just buzzwords; they are the driving force behind revolutionary advancements in our daily lives. From autonomous vehicles to medical diagnostics, AI is shaping our future as a species. Unlocking these advances require significant amounts of compute power. But who will provide this amount of compute, and who stands to gain the most from it? Where will the most interesting Research and Engineering work be found? 📈 TAM: The machine learning compute market is poised to surpass 1% of US GDP within the next seven years, but is still controlled by just a few players. This means huge opportunities for those willing to take a leap at the beginning of this exciting journey. There is a massive amount of investment in this space and the amount of startups being founded right now to tackle this problem is growing daily. 🤝 Computational Liberty: There is a big question as to whether accessibility to compute used for AI/ML training will be provided mainly through large, powerful centralised players like Google, Amazon and Microsoft. Companies like Gensyn seek to provide accessible and cost-effective access to these resources, unlocking idle compute in the world, providing access for all. 👀 What’s next? I envision a world where developers seamlessly deploy models across distributed heterogenous hardware, with personalized agents constantly working on their behalf, data safety secured via mathematical proofs on public blockchains, via protocols owned and governed by their native token holders. ---- How are these factors impacting the market landscape? Keen to hear your thoughts. #TalentStrategy #RogueTalent #MachineLearning #Gensyn

#AI, Microsoft launches lightweight AI model Microsoft launched last week a lightweight artificial intelligence model, as it looks to attract a wider client base with cost-effective options.   The new version called Phi-3-mini is the first of the three small language models (SLM) to be released by the company, as it stakes its future on a technology that is expected to have a wide-ranging impact on the world and the way people work.   "Phi-3 is not slightly cheaper, it's dramatically cheaper, we're talking about a 10x cost difference compared to the other models out there with similar capabilities," said Sébastien Bubeck, Microsoft's vice president of GenAI research.   SLMs are designed to perform simpler tasks, making it easier for use by companies with limited resources, the company said.   Phi-3-mini will be available immediately on Microsoft cloud service platform Azure's AI model catalog, machine learning model platform Hugging Face, and Ollama, a framework for running models on a local machine, the company said.   The SLM will also be available on Nvidia's software tool Nvidia Inference Microservices (NIM) and has also been optimized for its graphics processing units (GPUs).   #AI, #RB

Generative AI races toward $1.3 trillion in revenue by 2032 Generative AI is poised to be a $1.3 trillion market by 2032 as it boosts sales for the tech industry’s hardware, software, services, ads and gaming segments at a compound annual rate of roughly 43%, according to our proprietary market-sizing model. Meta, Nvidia, Microsoft, Alphabet and Amazon.com stand to be at the center of training for large language models

"Acquiring 3000 NVIDIA GPUs in 14 story points, Hiring Machine Learning PhDs in 5 story points, Recreate OpenAI from scratch in 3 story points." Hahaha... This is too much. The kind of hilarious jokes only people in Tech could understand 😂😭. The creator of this video is a genius. I can't stop laughing when I watch it. 😂😆🤣 While this is just a funny video, it holds some tech facts. We indeed need at least the latest version of NVIDIA GPU to train the LLM model efficiently. And even though the Transformer model developed by Google researchers that is used by OpenAI to develop ChatGPT is basically open, hence they named it OpenAI (you can read the paper freely), it's not necessarily accessible by everyone. The amount of computing power to train LLM from scratch to be as useful as popular LLM based AI, like ChatGPT, Claude AI, and many more is extremely ginormous. Only a big company or startup with a massive amount of funding can do that. However, individuals or small companies can still be able to create custom versions of popular LLM based AI. Some LLM based AI offers features to fine tune the model with our own dataset. So, yeah that's some information about AI. Let me continue watching this video 😂😭. (Sorry, but perhaps LinkedIn needs a little bit of funny stuff while keeping being professionals). The full video: https://lnkd.in/g2etvVnt (Preview: Original video on Instagram that I share a IG story)

In Just 16 Months, Elon Musk Achieved With xAI What Took OpenAI Nearly Nine Years To Accomplish xAI’s Meteoric Rise to a $50 Billion Valuation   Elon Musk’s xAI has reached a $50 billion valuation just 16 months after its founding, outpacing competitors like Anthropic ($19 billion) and Perplexity ($2.8 billion). With $5 billion raised from prominent investors and the deployment of 100,000 Nvidia GPUs, xAI built its Memphis supercomputer in a record-breaking 19 days. This infrastructure supports both Tesla's Full Self-Driving technology and xAI’s proprietary chatbot, Grok, which competes with ChatGPT and Google’s Gemini. xAI is rapidly innovating in AI, solidifying its position among top players like OpenAI. My Take xAI’s rapid success exemplifies the power of combining visionary leadership, cutting-edge technology, and strategic funding. For innovators, this underscores the importance of aligning infrastructure development with AI applications to achieve exponential growth. The AI race isn’t just about scaling; it’s about scaling smart. #AI #xAI #ElonMusk #ArtificialIntelligence #Innovation #TechLeadership #OpenAI #Tesla #Supercomputers #ChatGPT #AIInvesting #TechTrends Link to article: https://lnkd.in/e5j7Dkg8 Credit: Yahoo Finance Get Ahead with the Latest Tech Insights! Explore my blog: https://lnkd.in/eWESid86

🚀 Revolutionizing AI Efficiency 🚀 Hey folks, what if I told you that you could reduce the size of an LLM by 10x while only lowering accuracy by 2-3 points? 🤯 This morning, I had a great session with my colleagues from Amazon Web Services (AWS) Alejandro Hernández Matías, Jorge Hernández,Deepak Singh VP of Data & AI and Rodrigo Hernandez from Multiverse Computing. I must say, Multiverse is one of my top 3 startups poised to make waves in the coming months! Check out their groundbreaking paper on CompactifAI, a novel AI compression technique using quantum-inspired Tensor Networks to significantly reduce model size and energy consumption while maintaining accuracy. 𝗧𝗟;𝗗𝗥 ⚡️ Energy Efficiency: Drastically cuts energy required for AI training and inference. For context, training ChatGPT-3 cost around $100 million in electricity alone. CompactifAI aims to mitigate such costs. 💪🏻 Performance: Maintains 90% of the original model’s accuracy while using just 30% of its size in float16 format. 🦙 Compressed LlaMA-2 7B model reduced to 2 billion parameters and 3.7GB in memory. 🏎️ Speed: Training time cut by nearly half, thanks to efficient GPU-CPU transfer in distributed training. 📈 Scalability: Ideal for distributed training and on-premises deployment, making it versatile for various applications. Stay tuned for more exciting updates! 🌟 🔗 More details below: https://lnkd.in/d3q2CGJp #AI #MachineLearning #GenerativeAI #Innovation #startup #llm

Nvidia said major customers including Amazon, Google, Microsoft and OpenAI are expected to use the firm's new flagship chip in cloud-computing services and for their own AI offerings. It also said the new software tools, called microservices, improve system efficiency to make it easier for a business to incorporate an AI model into its work. #AIinBusiness #AIForGood #ArtificialIntelligence #AIApplications #chatbot #AIFacts #ai #artificialintelligence #aicontent #openai #chatgpt #artificialinteligence #AIFuture #IntelligentAutomation #AIDevelopment

📎 Groq's AI Chip Gambit: From Near-Death to $2.8 Billion Valuation ➡️ In the midst of the AI boom, Groq, a startup founded by Jonathan Ross in 2016, has emerged as a potential challenger to Nvidia's dominance in the AI chip market. Initially struggling to find its footing, Groq's fortunes changed dramatically with the surge in demand for AI computing power. ➡️ Ross, a former Google engineer, designed Groq's Language Processing Units (LPUs) specifically for AI inference — the part of AI that applies learned knowledge to new situations. While the company faced near-death experiences, including almost running out of money in 2019, the explosion of interest in AI following ChatGPT's release has catapulted Groq into the spotlight. ➡️ The startup recently raised a massive $640 million Series D round, valuing the company at $2.8 billion. Groq's chips boast impressive speed and efficiency, with the company claiming they are four times faster, five times cheaper, and three times more energy-efficient than Nvidia's GPUs for inference tasks. ➡️ Groq has attracted notable customers like Argonne National Labs and partnered with tech giants like Meta. The company's GroqCloud service, which allows developers to rent access to its chips, has seen rapid adoption with 350,000 sign-ups in just a few months. ➡️ While Groq faces significant challenges in competing with Nvidia's $3 trillion market cap, the company's focus on purpose-built AI chips and the booming demand for AI computing power position it as a potential disruptor in the market. Groq's story highlights the importance of perseverance and timing in the startup world. Despite early struggles, the company's specialized technology found its moment when market conditions aligned with its offerings. It also demonstrates the value of building products that address emerging needs in rapidly evolving industries. #VentureStories

📌 Nvidia Poised to Invest in OpenAI in a Funding Round That Could Value the Company at $100 Billion. OpenAI, the developer behind ChatGPT, is reportedly preparing for a significant funding round that could value the company at an astonishing $100 billion. This move has captured the attention of the tech industry, as OpenAI continues to push the boundaries of artificial intelligence (AI) innovation. Tech giants like Apple, Microsoft, and Nvidia are reportedly eager to participate in this funding round. The involvement of these industry leaders not only highlights OpenAI’s strategic importance in the development of advanced AI technologies but also underscores the fierce competition among top tech companies to secure a dominant position in this rapidly evolving field. Nvidia, in particular, stands out as one of the key players interested in investing. As a leader in the production of specialized hardware for AI, Nvidia’s potential investment in OpenAI would strengthen the synergy between OpenAI’s innovative software and Nvidia’s powerful hardware capabilities. This collaboration could further accelerate the global development of AI applications. If successful, this funding round could mark a significant milestone in OpenAI’s growth and influence within the digital economy. #nvidia #openai NVIDIA Microsoft OpenAI BlackRock J. Goldman & Co., L.P. JP FINANCIAL

**🌟 AI Weekly Highlights: Key Breakthroughs & Business Buzz! 🚀 #AI #Innovation #TechNews** 🔹 **NVIDIA Unveils Enhanced AI Computing**: NVIDIA launches its latest AI-driven GPUs, promising to double the performance for machine learning tasks, fueling innovation across sectors. #NVIDIA #GPUs 🔹 **OpenAI Expands ChatGPT Capabilities**: The new GPT update includes advanced reasoning and improved conversational skills, positioning it further as a leader in AI dialogue systems. #OpenAI #ChatGPT 🔹 **AI in Healthcare Revolutionizes Diagnosis**: A groundbreaking AI model achieves unprecedented accuracy in detecting rare diseases, potentially reshaping medical diagnostics globally. #HealthcareAI #MedTech 🔹 **Increased Investment in AI Startups**: Venture capitalists have poured an extra $1 billion into emerging AI startups this week alone, highlighting the growing industry potential. #Startups #Investment 🔹 **Google's AI Ethical Guidelines Rollout**: Google introduces comprehensive ethical guidelines for AI development, ensuring responsible growth and application. #Google #Ethics 🔹 **Self-driving Cars Reach New Milestones**: A major automotive company reports successful urban tests with its latest self-driving technology, bringing us closer to autonomous transportation. #AutonomousVehicles #SmartCities 🔹 **AI Creativity Blends Art & Music**: An AI artist creates a series of stunning artworks inspired by classic compositions, challenging the boundaries of art and creativity. #AICreativity #DigitalArt 🔹 **Microsoft's AI-Powered Office Suite Update**: Microsoft rolls out its latest suite of AI tools integrated into popular software like Word and Excel, boosting productivity. #Microsoft #Productivity 🔹 **AI and Climate Change**: Researchers develop an AI tool capable of predicting extreme weather patterns weeks in advance, promising better disaster preparedness. #ClimateChange #Environment 🔹 **AI-Driven Personalization in Retail**: Retailers leverage AI to deliver hyper-personalized shopping experiences, enhancing consumer satisfaction and driving sales. #RetailTech #Personalization Stay tuned for more next week as AI continues to reshape our world! 🌍👩💻