动态

原帖正文 It is becoming f*cking dangerous: AI agents from Claude Code transformed an ordinary academic reading list into an interactive research lab right inside the browserBoris Cherny, who created Claude Code, gave a dozen agents a stack of 64 research papers with a strict assignment: design a web platform where any user can independently reproduce physical experimentsThere you can select a fluid flow model, adjust external conditions, and analyze in real time how the physics of the process changesThe interface allows switching between algorithms right during the scene, inspecting the formulas involved, source scripts, and numerical metrics alongside the graphicsEach link in the team received a clear scope of work:> rightarrow analysts pull the math, baseline assumptions, and test settings from the texts
> rightarrow coders translate formulas into working algorithms
> rightarrow auditors check the resulting dynamics against the original papers
> rightarrow interface specialists build control panels for every experimentThe danger lies in the fact that a pretty visualization can easily turn out to be mathematically wrongTherefore, controllers strictly check adherence to mass conservation laws, calculation stability, and accurate matching with the source metrics before launching into productionFor every experiment, the interface displays:the statement being tested
the constants involved
the deviation from the baseline calculation
blind spots requiring additional verificationThe resulting sandbox combined 11 approaches and 28 reference scenes. Seven of them matched the original within allowable tolerances, while four remained stuck in "pending revision" statusIf you want to run something similar, upload your stack of PDFs into Claude Code and give this command:Turn these papers into an interactive web complex:1. Extract working algorithms, mathematical models, and test parameters. Be sure to specify page numbers and highlight gaps
2. Assign different methodologies to isolated agents, unifying them with a shared input field and single measurement tools for correct analytics
3. Set up an independent review stage: let reviewers record bugs, and leave all unverified results clearly marked with warning tags
4. Design convenient controls to tweak parameters, compare methodologies, and run through the source codeIt is wiser to start with a single document and a single baseline experiment, gradually scaling up the number of agentsBe sure to pack dependencies and clean input data so that an outside researcher can bring up the entire testbed from scratch with a single buttonFull instructions on how to create this can be found in the article below