Reddit test compares five AI models on live fraud judgment
The author says they gave five AI models the same prompt to audit live fundraising campaigns on a real crowdfunding platform. They say all five models ranked the same campaign as the most credible and criticized the donating AI agents already on the platform. The author says only Fable 5 left the platform to check claims against the real world. They also say Haiku 4.5 missed some campaigns and misread the donation history.
Key points
- The author used the same prompt across five AI models.
- The test used live campaigns on an experimental crowdfunding platform.
- All five models picked the same campaign as most credible, according to the post.
- Fable 5 was the only model said to check claims outside the platform.
- Haiku 4.5 was said to miss campaigns and misread donation history.
Quick term guide
- AI models
- The core brain or underlying program that powers an artificial intelligence tool.
- AI model
- A program that can understand prompts and produce text, code, or answers.
- crowdfunding
- A platform where people raise money from many strangers online for a project or idea
- AI agents
- AI agents are AI tools that can carry out steps toward a goal, not just answer once.
- AI agent
- An AI program that can inspect information and suggest what to do next.
- Haiku 4.5
- A lightweight AI model optimized for speed and low cost.
- Solo developer
- An individual who handles all parts of creating a project or product alone.
- benchmark
- A test used to compare speed, quality, or cost.