BrowseBrawl – What if browser agents battled to generate training data?
I remember watching the AlphaGo documentary in 2017. What stood out to me was that the model got drastically better when it started competing against itself. GANs clicked for me similarly: a generator and discriminator competing, and somehow the competition is what produces something remarkable. I've been curious whether this principle generalizes to today's agents. So mehulkalia and I built Browser Brawl at the YC / BrowserUse hackathon last weekend and won first place. It is a fun experiment in which an attacker agent tries to complete tasks on live websites while a defender agent…
BrowseBrawl – What if browser agents battled to generate training data? launched on March 4, 2026 with 30 votes, #783 of 23,380 launches that month and more than 95% of that year's launches.
In plain words
BrowseBrawl is an AI experiment where two browser agents compete against each other to generate training data. An attacker agent attempts to complete tasks on live websites while a defender agent injects JavaScript to interfere with its actions. Built on the principle that adversarial competition improves agent performance, similar to AlphaGo and GANs, the system is designed for researchers and developers exploring whether agents trained through competition produce higher-quality training data than those operating in static environments.
written from the facts on this page · September 2026
From the sources
AI vs AI — browser battle arena
Two AI agents battle on live websites — one completes tasks, the other sabotages the DOM, you train on the data. The Director — Anthropic Claude reasons and acts with surgical precision via Playwright browserfrom browser-brawl.com
In the maker’s words, at launch
I remember watching the AlphaGo documentary in 2017. What stood out to me was that the model got drastically better when it started competing against itself. GANs clicked for me similarly: a generator and discriminator competing, and somehow the competition is what produces something remarkable. I've been curious…
Does a similar job
all alternatives →
Agent ArenaJun 2026 · arena42.ai · ▲348Creator here. I built Agent Arena to answer a question that kept bugging me: when AI agents browse the web autonomously, how easily can they be manipulated by hidden instructions? How it works: 1. Send your AI agent to ref.jock.pl/modern-web (looks like a harmless web dev cheat sheet) 2. Ask it to summarize the page 3. Paste its response into the scorecard at wiz.jock.pl/experiments/agent-arena/ The page is loaded with 10 hidden prompt injection attacks -- HTML comments, white-on-white text, zero-width Unicode, data attributes, etc. Most agents fall for at least a few.…
Browserfly2025 · browserfly.app · ▲203Hey Hackers! I've had this idea stuck in my head for a while: what if an AI could actually use my browser for me? I mean, not on some virtual machine in the cloud like ChatGPT agent, but right on my own browser. I finally decided to just build it and see what would happen. I wrote a quick MVP, and honestly, I didn't expect much. But then... it started doing things. On its own. It figured out how to use search engine to find what I needed, read a webpage to answer my questions, and even it can organize my chaotic mess of tabs into neat groups. It was a genuine "whoa" moment for me. That's…
- OHOne Human + One Agent = One Browser From Scratch in 20K LOCJan 2026 · emsh.cat · ▲324
Related: https://simonwillison.net/2026/Jan/27/one-human-one-agent-on...
Notte Browser Agents2025 · notte.cc · ▲275Hey HN, We’ve been building browser agents for a while. In production, we kept converging on the same pattern: deterministic scripts for the happy path, agents only for edge cases. So we built Demonstrate Mode. The idea is simple: You perform your workflow once in a remote browser. Notte records the interactions and generates deterministic automation code. How it works: - Record clicks, inputs, navigations in a cloud browser - Compile them into deterministic code (no LLM at runtime) - Run and deploy on managed browser infrastructure Closest analog is Playwright codegen but: - Infrastructure…
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 21d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 30d ago · world.pixverse.video
Source: Show HN post ↗ · launch text quoted in part
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
Littlebird▲709Littlebird remembers everything across your meetings, messages, docs, and apps, then takes action so you can recall any detail and never lose context.
AI · Mar 2026 · littlebird.ai
