Skip to content

README.md

Latest commit

 

History

History
92 lines (62 loc) · 2.72 KB

README.md

File metadata and controls

92 lines (62 loc) · 2.72 KB

RecoverAI

A satirical Spring Boot application that helps organizations recover from "AI psychosis" - the overenthusiastic and reckless adoption of AI technologies.

How It Works

Run the diagnose command in the interactive shell to start a guided questionnaire. The application will:

  1. Intake: Collect symptoms across three categories (Delusion of Velocity, Hallucinatory Resource Planning, Reality Detachment)
  2. Diagnosis: Evaluate the stage of AI psychosis (Prodromal, Acute, or Residual)
  3. Recovery Planning: Generate a recovery strategy tailored to the affected discipline
  4. Diplomatic Communication: Produce a script to address the issue with leadership

Symptom Categories

Category Description
Delusion of Velocity Obsession with AI-generated output metrics (PRs, lines of code)
Hallucinatory Resource Planning Belief that AI replaces human expertise and capacity
Reality Detachment Disconnection from actual costs, outcomes, and accountability

Stages of AI Psychosis

Stage Description
Prodromal Unusual interest, AI-first hype, mild detachment from reality
Acute Hallucinations. Managers pushing AI code to prod without PRs
Residual The crash. Tech debt, broken production, team burnout/cynicism

Supported Disciplines

Engineering, Product, Dev Management, Leadership, Marketing, Finance, and HR

Requirements

  • Java 21
  • Maven

Configuration

Set the following environment variables for LLM configuration:

OPENAI_CUSTOM_BASE_URL=https://api.example.com
OPENAI_CUSTOM_MODELS=model-name
OPENAI_CUSTOM_API_KEY=your_api_key

Optionally, set OPENAI_CUSTOM_COMPLETIONS_PATH if the completions endpoint differs from the default OpenAI path (/v1/chat/completions):

OPENAI_CUSTOM_COMPLETIONS_PATH=/custom/chat/completions

Building

mvn clean package

Running

java -jar target/RecoverAI-0.0.1-SNAPSHOT.jar

Or with Maven:

mvn spring-boot:run

Once running, use the diagnose command to start the assessment.

Available Commands

Command Description
diagnose Run the full diagnostic workflow
memoryFetch <query> Search memory notes using vector similarity
memoryWrite <text> Store text in memory for later retrieval

Technology Stack

  • Spring Boot 3.5
  • Embabel Agent 0.3.4 - AI agent framework
  • Archiledger - Knowledge graph builder using memory notes, where the LLM classifies AI psychosis analyses
  • Spring Shell - Interactive CLI

License

See LICENSE file for details.