--- title: "My Favorite Uses of AI" description: "Discover how AI transforms daily workflows, from book selection to email management. A VC's practical guide to leveraging AI tools for better productivity and decision-making." categories: ["trends","AI"] keywords: ["AI productivity tools","Gemini AI","Claude AI","AI summarization","voice recognition AI","venture capital","Tomasz Tunguz","academic paper analysis","email management AI"] ai_summary: "Explore innovative ways to leverage AI for productivity, from summarizing emails to managing academic papers." date: 2024-04-14 lastmod: 2026-07-23 canonical_url: https://www.tomtunguz.com/my-favorite-uses-of-ai/ author: "Tomasz Tunguz" --- Recently, I was at the library & stood in front of a book that might be a good read. I fired up Gemini, asked it for a plot summary & a digest of the reviews before deciding to read it. So I started a list : what are the ways I use AI that are new in the last year? ![image](https://res.cloudinary.com/dzawgnnlr/image/upload/ewgectojccjz9aquotcy.png) I often publish tables within these blog posts. I write posts in [markdown](https://www.markdownguide.org/extended-syntax/), a language that requires making tables in a unique format that looks like this | Column 1 | Column 2 | | 2021 | 140m | There's a nifty little plugin for my text editor called [Gen](https://github.com/David-Kunz/gen.nvim) that converts data in nearly any format into these tables. I type comma+o & Gen formats the table in a few seconds. Or correct the grammar in a sentence or execute a quick search on a topic or summarize an email or format a list.
Within my email editor, I summarize long emails with the same command. Also, long articles on the internet. AI is driving a broader behavior change : summarize, then determine whether to invest the effort to read. I upload academic papers to a large language model, read the conclusion, & then badger the AI with questions, especially [comparative ones](https://tomtunguz.com/bcg_ai_paper/). Recently, Claude helped me understand the differences between the transformer AI architecture & [Google's Griffin architecture](https://arxiv.org/abs/2402.19427). Here's another example exploring the [Memory Wall](https://medium.com/riselab/ai-and-memory-wall-2cb4265cb0b8) within AI. ![image](https://res.cloudinary.com/dzawgnnlr/image/upload/mgub2p1pks9fpheospgq.png) Voice is my most frequently used AI. I dictated to the Google Android [Recorder app to dump all my thoughts at once](https://video.twimg.com/tweet_video/GAsKvSjWwAALLB9.mp4 ) and then click the summarize button to send myself notes or todos. [![image](https://res.cloudinary.com/dzawgnnlr/image/upload/trmiwhi3wiyefopguett.png)](https://twitter.com/i/status/1732490330941845539) On my Mac, it's the same thing with [Whisper](https://github.com/ggerganov/whisper.cpp). I wrote a [Python script](https://gist.github.com/ttunguz/686bb1e28f2df4e66373cdf57bf61627) so I can hold down the left option key on my Mac, & the computer will transcribe what I say while it's depressed. I use the same software to trancribe & summarize podcasts with [Ollama](https://ollama.com/) & an LLM locally. Image generation for blog post images & code completion remain persistent uses as well. What's your favorite way to use AI?