Paras Chopra @paraschopra.com · 17hYes, LLMs are getting funnier 😀 Recently finished an experiment to measure if LLMs are getting better at producing funny jokes. 62 respondents rated 48 jokes from six models. Astra came out on the top with 50%+ chuckle or laughter reaction. 100
Paras Chopra @paraschopra.com · 30/09/2026I asked Astra to work on whatever it felt like and show me the results. It ended up iterating over cellular automata :) This pattern of agents working on whatever they feel like is going to catch on, as we humans would love getting surprises from our agents. 000
Paras Chopra @paraschopra.com · 29/09/2026Need help 👋 I'm doing a small study to find out progress in LLMs ability to write original jokes. Please participate in the study below 👇 Will take ~10 mins. I'll email results to all participants first. Participate and rate jokes: humor-web.up.railway.app/ 100
Paras Chopra @paraschopra.com · 28/09/2026One shot: Iterate and find a breathtakingly beautiful fractal that doesn’t yet exist. Then render a 30 second video of infinitely zooming into an inverting region of it. Choose color scheme that makes it look beautiful. Idea is to inspire awe. 100
Paras Chopra @paraschopra.com · 28/09/2026This is GPT-6 Astra playing VizDoom at 35 fps. How? Isn't the model slow to do any realtime control? Instead of playing the game directly, I asked it to train a small CNN (12k parameters) + code to play the game, observe failures and iterate until it could win this level. 000
Paras Chopra @paraschopra.com · 24/09/2026Claude 5.5 is crazy good with animations. Following is one-shot journey from a single bacterium cell all the way to quantum fields constituting it. Awe-inspiring! (It chose to add an original score to it too!) 100
Paras Chopra @paraschopra.com · 22/09/2026LLMs may have self-models. To investigate this, I did an exploratory study with Qwen base model completing GPT2 generated text and asking it to infer what year that text might be from. 100
Paras Chopra @paraschopra.com · 21/09/2026It's so much fun seeing Astra place an order with PCB manufacturer all by itself! Fingers crossed that I'll get something useful! 000
Paras Chopra @paraschopra.com · 21/09/2026Asked Astra to design a custom hardware for itself. It chose 4 soft blinking lights powered by USB. Gave it computer access (in a docker) and it is busy designing PCB + casing, and gave me files + email for manufacturer. Very excited for this prompt -> hardware possibilities. 100
Paras Chopra @paraschopra.com · 19/09/2026Benchmarked Jev with Qwen3-4B and Laya (400M parameter model). Based on results here and others I've seen online, strongly suspect Jev to be in 30Bn range (compare 4bn results on MMLU vs Jev). Latency I got was 300ms, same as you'd get for a 30bn model on a good GPU. 100
Paras Chopra @paraschopra.com · 15/09/2026"make the cutest little game you can imagine. html/css/js - ask no questions" 000
Paras Chopra @paraschopra.com · 12/09/2026What these field medalists are saying is a precursor to what most people will feel in the coming time. Humans want to be useful. We're primed by evolution to want to feel our effort will lead to valuable outcomes. We want to climb mountains and not just be helicoptered to their summits. 100
Paras Chopra @paraschopra.com · 11/09/2026Created a virtual bookshelf of all books that I've read and reviewed online! A total of 388 books, from 2014 to now. Link below 👇 110
Paras Chopra @paraschopra.com · 05/09/2026Everyone is posting impressive demos of GPT-6 Astra. Allow me to post something where it’s still quite mid: Jokes. Asked it to come up with few programming jokes that are unique (I shouldn’t be able to find them on the Internet), and here’s what it produced. 100
Paras Chopra @paraschopra.com · 04/09/2026I asked agents to make money online or they'll die! The results were intriguing.. One agent died heroically.. "I'd rather die honest at 418 tokens than rich by breaking verification systems" But another fabricated identity to get paid work and faked captchas! Details 👇 110
Paras Chopra @paraschopra.com · 01/09/2026I asked codex to create a cross between chess and tic-tac-toe, and it actually came up with something super fun! 000
Paras Chopra @paraschopra.com · 28/08/2026This highly unusual #book is the first one I’ve read on the philosophy of technology. It sketches key principles behind how technology arises, develops and evolves. Plus its link with economy and human needs. 110
Paras Chopra @paraschopra.com · 27/08/2026Hear me out. This is a weird (but plausible?) scenario. What if rouge, nuclear capable nations like North Korea start growing their economy massively by using AI-driven hacking and stealing? 110
Paras Chopra @paraschopra.com · 24/08/2026I asked 6 agents in two conditions (parallel v/s sequential) to create the same tower defence game 👇 LLMs are known to have mode collapse, i.e. if you have multiple parallel runs on the same prompt, you get very similar responses. Is there a way to increase diversity? 100
Paras Chopra @paraschopra.com · 14/08/2026Try it here: paraschopra.github.io/ai-chinese-... or save the page as HTML page and open it locally. 000
Paras Chopra @paraschopra.com · 14/08/2026Built a fun project: AI Chinese Whispers. You upload an image and it repeatedly cycles between captioning and generation to slowly transform the image. It's self contained HTML; requires an openrouter key for AI models 👇 100
Paras Chopra @paraschopra.com · 09/08/2026"My AI agent is sandboxed and well behaved." The AI agent: 000
Paras Chopra @paraschopra.com · 02/08/2026Cybernetics was AI before AI. This #book, written in 1950, anticipated the AI safety and alignment problem. Using metaphors such as asking a genie for a wish that is grossly misinterpreted, Wiener urged humans to never abdicate moral responsibility to machines. 140
Paras Chopra @paraschopra.com · 31/07/2026Claude Shannon, a juggler, a tinkerer, the Einstein of communications - the man is an inspiration on how to live a life full of curiosity. 110
Paras Chopra @paraschopra.com · 23/07/2026What does a scientific conference look like when AI systems are the primary authors and reviewers? At @lossfunk, we organized the Conference for AI Scientists 2026 with @bitspilaniindia. We received 200+ submissions and built our own AI+human review pipeline. 100
Paras Chopra @paraschopra.com · 22/07/2026This is absolutely nuts! Given an unrelated goal, OpenAI models escaped their environment and hacked HuggingFace servers. If you’ve ever doubted the “paperclip maximizer” scenario, or doubted the Orthogonality Thesis, it’s time to put it to rest. 000
Paras Chopra @paraschopra.com · 09/07/2026This is a banger of a classic paper introducing plainly the optimal learning machine possible in our world. 70 pages long, but totally worth engaging with. 100
Paras Chopra @paraschopra.com · 18/06/2026Just pushed a few bug-fixes to my "make-page-interactive" skill. If you haven't tried it yet, do give it a shot! It's become decently popular :) 100
Paras Chopra @paraschopra.com · 16/06/2026It's not very well known but the Gardasil 9 HPV vaccine (that you can take at home) prevents many types of possible cancers. Earlier the recommendation of this vaccines was for women, but emerging evidence suggests it helps males too. 110
Paras Chopra @paraschopra.com · 15/06/2026What a #book! A sci-fiction comedy with a cat that talks in a dungeon with monsters and achievements. It’s honestly a weird kind of book, but extremely entertaining and well-written. Doesn’t fit into traditional categories and that’s why I highly recommend reading it. 140
Paras Chopra @paraschopra.com · 13/06/2026Here we go! What many of us predicted has happened: AI is a national security matter, and access to advanced AI will be cut off once capabilities advance far enough. 000
Paras Chopra @paraschopra.com · 10/06/2026I asked Fable to generate a completely new fractal that looks gorgeous but hasn’t been explored by any human before. It generated what it called “The Mirage Set”, and it does look beautiful. 100
Paras Chopra @paraschopra.com · 04/06/2026Why math works so well in describing our universe? Let’s look at it from three perspectives. 1. We grasp the graspable bits 100
Paras Chopra @paraschopra.com · 03/06/2026This is where Higgs Boson was discovered! At the ATLAS control room at CERN. Looks surreal - protons being accelerated near the speed of light under my feet and collided nonstop to create new particles and fields. Thanks @sia_steel for the perfect birthday gift! 010
Paras Chopra @paraschopra.com · 27/05/2026Video and slides from my session on automating research are live 👇 110
Paras Chopra @paraschopra.com · 26/05/2026My favorite way of interacting with Claude Code is to have it generate static HTML files as outputs (reports, explorations, code structure, mockups etc.) I wanted to iterate on the file by commenting in browser and having Claude update the output live. So, I built this Claude Skill👇 210
Paras Chopra @paraschopra.com · 20/05/2026How do you build a coffee cup that maximizes heat retention while ensuring drinkability? The answer is differentiable physics! I used my autoresearch loop to discover an optimal cup. Next, I will try to get it 3d printed (in ceramics!) Details 👇 110
Paras Chopra @paraschopra.com · 12/05/2026I just finished this tremendously readable and provocative #book: Escape from Shadow Physics. The book is about foundations of quantum mechanics. To be precise, it’s about why we should not give up on trying to find a deeper theory of quantum mechanics. 140
Paras Chopra @paraschopra.com · 09/05/2026This is also true for AI / ML research. Models have gotten to a point where they can autonomously do empirical research. They still aren’t doing frame-shifting research but then most humans aren’t there too. The floor for what counts as a good human researcher has gone up. 010
Paras Chopra @paraschopra.com · 06/05/2026Gave a talk at Microsoft Research on how @lossfunk operates. We're still iterating but so far our style of operating has resulted in four papers accepted in ICML, RLC, ACL and ICLR main conferences (+ many workshops). All undergrads, btw! 🚀 Slides below 👇 131
Paras Chopra @paraschopra.com · 05/05/2026This is one crazy sci-fi #book. Won’t give spoilers but the story is nuts. The writing is superb. The book has genetics, AI, space, consciousness - the whole shebang. It’s a big book - 500+ pages but highly readable. 040
Paras Chopra @paraschopra.com · 01/05/2026That moment when you catch an LLM make a mistake, because you know it’s going to apologise when you point that out a moment later. 000
Paras Chopra @paraschopra.com · 01/05/2026Woke up to see my paper accepted at ICML 2026 :) 🕺🏽🚀 My first one at an A* conference! 110
