• AI KATANA
  • Posts
  • Amazon Developing Custom AI Processors To Compete With NVIDIA

Amazon Developing Custom AI Processors To Compete With NVIDIA

Also: How Gen AI Is Already Impacting the Labor Market

Author

AI KATANA
November 12, 2024

In today’s updates, Amazon steps up its game in the hardware space with custom AI processors designed to reduce reliance on NVIDIA and enhance its AI ecosystem. Meanwhile, new research from HBR highlights the shifting job landscape driven by generative AI, with some creative roles in decline while demand for AI proficiency rises. On the marketing front, AI is transforming ad targeting, enabling precision personalization that boosts effectiveness across campaigns. Plus, an insightful hands-on with Meta’s AI-powered Ray-Ban glasses reveals both the perks and limitations of wearable tech for daily tasks. Google’s latest AI “learning companion” aims to elevate chatbot interactions by offering educational depth, and Deckmatch secures funding to refine its VC matching platform with advanced AI capabilities.

Sliced just for you:

  • 🖥️ Amazon Developing Custom AI Processors To Compete With NVIDIA

  • 📉 HBR Research: How Gen AI Is Already Impacting the Labor Market

  • 🎯 AI is taking ad targeting to a new level

  • 👓 I went shopping wearing the AI-powered Meta Ray-Bans — here’s what I love and what annoyed me

Amazon is developing custom AI processors through its Annapurna Labs, aiming to reduce its dependency on NVIDIA’s GPUs and cut operational costs. With high demand and premium pricing for NVIDIA’s market-leading GPUs, Amazon’s move to create its own chips reflects a broader industry shift toward in-house solutions for AI infrastructure. Amazon’s custom chips, including the Trainium processors introduced in 2023, are being utilized by partners like Anthropic, which leverages these for its Claude AI model. These new chips, built in collaboration with Taiwan’s Alchip and manufactured by TSMC, enable Amazon to support a growing ecosystem of over 50,000 AWS customers with dedicated hardware solutions. This development places Amazon alongside other tech giants like Alphabet and Meta, which are also investing in proprietary AI processors to maintain competitiveness in the AI space.

📉 HBR Research: How Gen AI Is Already Impacting the Labor Market

Research on generative AI’s impact reveals significant changes in the labor market, particularly in automation-prone fields like writing, coding, and graphic design. Analyzing 1.4 million job postings on a freelancing platform from 2021 to 2023, researchers found a notable decline in demand for these roles following the release of tools like ChatGPT and image generators (e.g., DALL-E). Writing jobs saw a 30% drop, while graphic design postings decreased by 17% within a year. This shift has increased competition among freelancers, raising the number of bids per job, especially in fields susceptible to automation. Employers now seek more complex skills and are willing to pay higher rates for these automation-prone roles. Furthermore, the demand for AI-related skills is rising, with many job posts specifically listing ChatGPT proficiency. Researchers recommend proactive upskilling and reskilling to help workers adapt to these new AI-driven demands, emphasizing that businesses should strategically integrate AI to complement rather than replace human talent, positioning organizations for sustained growth in an AI-enhanced job market.

🎯 AI is taking ad targeting to a new level

AI is revolutionizing advertising by enhancing the precision and personalization of ad targeting, enabling marketers to reach audiences in novel and impactful ways. According to Kern Schireson, CEO of Known, AI allows advertisers to optimize messaging by determining the ideal timing, context, and platform for each audience, a task that would otherwise be unmanageable manually. Known’s innovative approach includes tuning ads to individual viewers, as demonstrated in a Super Bowl campaign featuring John Stamos, where AI customized the ad to mention the viewer’s location. Additionally, Known’s AI chatbot, “The Big Letbotski,” effectively mined 80,000 subreddits to identify relevant discussions for Shake Shack’s chicken sandwich promotion, resulting in a significant sales boost. By integrating these advanced tools, AI enables a new level of tailored ad experiences, enhancing relevance and effectiveness across various platforms.

The AI-powered Meta Ray-Ban glasses combine classic eyewear design with modern technology, offering features like voice-command texting, music streaming, video capture, and real-time data access. In a practical shopping scenario, they allowed for hands-free note-taking, seamless text communication, and real-time queries about products, making errands more efficient and enjoyable. The glasses also enable direct photo and video capture, which helped the user avoid impulsive purchases by visually tracking items of interest. Integrated with Meta platforms, they support live-streaming, allowing users to share experiences instantly on social media. Despite impressive functionality, the glasses have limitations, including a relatively short battery life (about four hours) and the need for Wi-Fi, which may limit usage in certain settings. Despite these drawbacks, the Meta Ray-Bans offer a streamlined and engaging experience for routine tasks, enhancing productivity and connectivity on the go.

🛠️ AI tools updates

Google’s new experimental AI tool, Learn About, is designed as a “learning companion” that goes beyond typical chatbot responses. Built on Google’s LearnLM model, this AI tool provides responses with educational structure, interactive visuals, and contextual elements to aid learning. Unlike Google’s Gemini, which displays brief, fact-based summaries with supporting links, Learn About emphasizes a richer learning experience, offering additional context like definitions, why topics matter, and suggestions for further exploration. For example, it includes textbook-style explanations and vocabulary aids that make complex topics more accessible. This approach reflects Google’s efforts to tailor AI to educational needs, creating a more structured and interactive way for users to understand complex information.

💵 Venture Capital updates

Deckmatch, an AI-focused startup based in Oslo, recently raised $3.1 million in Seed funding to enhance its AI-driven platform for matching venture capitalists (VCs) with startups. Founded in 2023 by Leo Gasteen and Walid Mustapha, Deckmatch automates private market workflows through a suite of REST APIs that convert pitch decks into structured data, integrating seamlessly with CRMs like Affinity and Slack. This capability allows VCs to process pitch decks rapidly, streamlining decision-making. Alongside the funding, Deckmatch launched a beta version of AlphaLens, an AI-powered product-centric search engine that offers comprehensive insights into market products, shifting the focus from firm data to product-level intelligence. Backed by Alliance VC, Luminar Ventures, and others who were also early customers, Deckmatch is expanding into the U.S. with an office in Palo Alto, aiming to redefine how private markets evaluate innovation.

🫡 Meme of the day

⭐️ Generative AI image of the day