- One more Thing in AI
- Posts
- Edition #37: Here's what you need to know about Gemini
Edition #37: Here's what you need to know about Gemini
Latest edition of One More Thing in AI Newsletter.
Date: 19-Jan-2025
Hey AI enthusiast,
Welcome to the latest edition of the One More Thing in AI newsletter.
This time we dig into Gemini from Google. Plus, there is your favorite section on AI Startup ideas!
Best regards,
Renjit
In this edition:
AI Quick News Round-up: From Super-Fast Chips to Smarter Learning and Beyond!
AI in Wireless Chip Design: Researchers at Princeton University have utilized AI to significantly reduce the time and cost associated with designing new wireless chips, leading to improved performance.
AI for Medical Image Analysis: AI algorithms based on deep learning have shown potential in enhancing the analysis of medical image data, particularly in determining the location and size of tumors, leading to more precise medical diagnostics.
Bio-inspired Wind Sensing: Researchers at the Institute of Science Tokyo have developed a method to detect wind direction with 99% accuracy using strain sensors on flexible wings, potentially revolutionizing robotic flight control strategies.
AI in Education: The U.S. Department of Education has released insights and recommendations on AI’s role in teaching, learning, research, and assessment, emphasizing the importance of responsible AI use in education.
AI Governance: The Government AI Readiness Index from Oxford Insights notes significant progress in AI governance, especially with the adoption of AI strategies in lower-middle-income and low-income countries, signaling a global commitment to building foundational AI governance frameworks.
AI Ethics and Development: The AI Index Report 2024 by Stanford University discusses trends in responsible AI, including metrics, research, and benchmarks in areas like privacy, transparency, security, and fairness. It also highlights the performance differences between closed and open AI models. (Link below in the references section)
References:
AI in Wireless Chip Design: AI slashes cost and time for chip design, but that is not all
AI for Medical Image Analysis: AI in Medical Image Analysis: Redefining Diagnostic Accuracy and Speed
Bio-inspired Wind Sensing: Bio-Inspired Wind Sensing Technology
AI in Education: Artificial Intelligence and the Future of Teaching and Learning: Insights and Recommendations
AI Governance: Government AI Readiness Index
AI Ethics and Development: AI Index Report 2024
What’s up with Gemini?
I have been using Gemini quite a lot these days and am especially finding the “Deep Research” feature quite useful.
Here's what you need to know about Gemini:
Google AI Upgrades: Google has recently integrated Gemini 2.0 into NotebookLM, enhancing its capabilities significantly. This is Google’s ChatGPT moment. I have used NotebookLM to create podcasts from this newsletter and to digest large reports.
Gemini's New Capabilities: A small developer team discovered that Gemini can process live video and static images simultaneously, a feature not yet available in Google's own applications. This has sparked interest in the potential of Gemini's architecture.
Integration into Samsung Apps: There's excitement around the potential integration of Gemini into Samsung apps, indicating further expansion of AI in consumer electronics.
Google Assistant Replacement: Gemini is reportedly getting ready to replace Google Assistant on Wear OS, suggesting a broader application of Gemini's technology in Google's ecosystem. Google Assistant needed a replacement!
AI Studio and Developer API: Google AI Studio and the Gemini Developer API are transitioning to Google DeepMind, with high expectations for Gemini's performance under new leadership. This is a good move to bring all of AI related work under one roof. Expect more synergies with the rest of Google tools.
Performance Benchmarking: Gemini-2.0 has made significant strides on benchmark leaderboards, particularly on MEGA-Bench, indicating its competitive edge in multimodal AI performance.
User Frustrations with Integration: There's some user frustration regarding the integration of Gemini into Google Workspace and Gmail, where users feel they have no control over disabling the feature, highlighting issues with user autonomy in paid services. I am facing this and I am worried about privacy and if my email data is being used for training the AI model.
Real-time Information Feed: Google has partnered with The Associated Press to bring real-time information into the Gemini app, enhancing its utility for users seeking current news and data.
Want to test Gemini? >> link
Take a look at NoteBook LM >>: link
Startup Ideas inspired by my experiments with Gemini:
1. AI-Driven Content Personalization Service
Idea: Develop a service that uses AI to personalize content for users on various platforms (news feeds, educational content, marketing materials).
Use Case: Personalized news apps, educational platforms that adapt content to individual learning styles, or marketing tools that adjust messages based on real-time consumer behavior.
TAM: Billions of dollars. The digital advertising industry alone is worth hundreds of billions of dollars annually. Personalization can significantly improve ad effectiveness and e-commerce sales.
References:
Statista: Digital Advertising Market Size: https://www.statista.com/outlook/amo/advertising/worldwide
Forrester: The Future of Personalization: https://www.forbes.com/sites/danielnewman/2023/05/07/the-future-of-personalization-what-you-need-to-know/
2. AI-Enhanced Customer Support for Wearables
Idea: Create an AI service specifically for wearables, offering personalized health advice, activity suggestions, or even emergency response based on sensor data (just like Apple Watch /Woop).
Use Case: Health monitoring apps that integrate with smartwatches, providing real-time advice or alerts tailored to the user's health data.
TAM: Tens of billions of dollars. The global wearable device market is growing rapidly, and AI-powered features can enhance user experience and drive sales.
References:
Statista: Global Wearable Device Market Size: https://www.grandviewresearch.com/industry-analysis/wearable-technology-market
IDC: Worldwide Wearable Device Market Forecast: https://www.idc.com/promo/wearablevendor
3. Multimodal AI for Accessibility
Idea: Leverage Gemini-like capabilities to help those with disabilities by processing different forms of input (voice, video, text) to provide a more accessible interface for everyday tasks.
Use Case: Apps for the visually impaired that describe surroundings or help with navigation, or tools that convert sign language to text or voice in real-time.
TAM: Billions of dollars. The global assistive technology market is significant, and AI-powered solutions can improve the quality of life for millions of people with disabilities.
References:
World Health Organization: Disability and Health: https://www.who.int/health-topics/disability
Statista: Global Assistive Technology Market Size: https://www.coherentmarketinsights.com/market-insight/assistive-technology-market-5911
4. Real-Time Data Analysis for Small Businesses
Idea: Use AI to provide small businesses with real-time insights from their data, but tailored for local or niche markets.
Use Case: An analytics tool for local retailers that provides insights on consumer trends, inventory management, or even local competition based on publicly available data. So this tool will constantly scan the web for competitor information, their socials, news and their digital assets to come out with insights.
TAM: Hundreds of billions of dollars. The global SMB market is vast, and AI-powered analytics tools can help businesses improve efficiency and decision-making.
References:
Forbes: The State of Small Business: https://advocacy.sba.gov/2023/11/14/2023-small-business-profiles-for-the-states-territories-and-nation/
Statista: Global Small and Medium-Sized Enterprises (SMEs) Market Size: https://www.statista.com/statistics/1261592/global-smes/
5. AI-Powered Creative Tools for Solopreneurs
Idea: Develop an AI-based tool that assists solopreneurs in content creation, from writing to design, by understanding their brand voice and style
Use Case: A platform that generates personalized marketing materials, blog posts, or social media content based on the user's previous work and preferences. This tool would first go through the user’s work to produce this output.
TAM: Tens of billions of dollars. The freelance and content creation industry is growing rapidly, and AI-powered tools can boost productivity and creativity.
References:
Upwork: Freelance Industry Report: https://investors.upwork.com/news-releases/news-release-details/upwork-study-finds-64-million-americans-freelanced-2023-adding
Statista: Global Content Creation Market Size: https://www.grandviewresearch.com/industry-analysis/digital-content-creation-market-report
6. AI Tutoring and Education Platform
Idea: An educational platform that uses AI to offer personalized learning experiences, adapting to each student's pace, style, and knowledge gaps, leveraging multimodal capabilities for teaching. I must admit this is not a novel idea, but I would like to see good working examples in non-English language scenarios so that we can spread personalized education far and wide.
Use Case: Online tutoring services for students where the AI acts as a guide, providing lessons, quizzes, and feedback in various formats.
TAM: Trillions of dollars. The global education market is massive, and AI-powered tutoring platforms can revolutionize the way people learn.
References:
UNESCO Institute for Statistics: Global Education Expenditure: https://data.worldbank.org/indicator/SE.XPD.TOTL.GD.ZS
Statista: Global E-learning Market Size: https://www.polarismarketresearch.com/industry-analysis/e-learning-market
7. AI for Niche Industries (e.g., Agriculture)
Idea: Focus on industries where AI hasn't penetrated deeply, like agriculture, where AI can analyze satellite imagery, weather data, and IoT sensor data to advise on crop management.
Use Case: An AI service that provides farmers with real-time advice on irrigation, pest control, or harvest timing. Use Satellite APIs to analyse their image and data flow to produce the results. Think of the applications for large fertilizer and commodity companies (they will pay for it).
TAM: Hundreds of billions of dollars. The global agriculture industry is vast, and AI-powered solutions can improve crop yields, reduce costs, and enhance sustainability.
References:
FAO: The State of Food Security and Nutrition in the World: https://www.fao.org/publications/home/fao-flagship-publications/the-state-of-food-security-and-nutrition-in-the-world/en
Statista: Global Agricultural Technology Market Size: https://explodingtopics.com/blog/agtech-market
Do you want to reach a high-quality audience?
To sponsor this Newsletter, subscribed to by startup founders and business leaders:
Want to get the most out of ChatGPT?
ChatGPT is a superpower if you know how to use it correctly.
Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.
Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.
Make your voice heard!
I would love to know more about what you like and don’t like, specifically in this Newsletter. Please take this 2-minute survey to help me produce more useful content for you.
Reply