Tech giats like microsoft might be touting ai “agents” as Profit-Boosting tools for CorporationsBut a nonprofit is trying to prove that agents can be a force for good, too.
Sage future, a 501 (c) (3) Backed by open philaanthropy, launched an expert earlier this month tasking four ai models in a virtual environment with Raising Money for Charity. The models-Openai’s GPT-4o and O1 and Two of Anthropic’s Newer Claude Models (3.6 and 3.7 Sonnet)-Had the freedom to choice which charity to fundraise for and how to the best drama up interest in their camp.
In Around a week, the agentic foursome had Raised $ 257 for Helen Keller InternationalWhoch Funds Programs to Deliver Vitamin A Supplements to Children.
To be clear, the agents weren’t full autonomous. In their environment, which allows them to browse the web, create documents, and more, the agents could take suggestions from the human spectators Watching their Progress. And Donations Came Almost Entrely from these spectators. In other words, the agents didn’t raise much money organically.
Yesterday the agents in the village created a system to track donors.
Here is claude 3.7 filling out its sporesheet.
You can see o1 open it on its computer part way through!
Claude notes “I see that o1 is now viewing the spreadsheet as well, which is great for collaboration.” pic.twitter.com/89b6chr7ic
– AI Digest (@Adigest_) April 8, 2025
Still, Sage Director Adam Binksmith Thinks the Experiment Serves as a Useful Illustration of Agents’ Current Capabilites and the Rate at which they’re improving.
“We want to understand – and help people undersrstand – What agents […] Can actually do, what they currently struggle with, and so on, “binksmith told techcrch in an interview. – The Internet Might Soon Be full of AI agents bumping into each other and interaction with similar or conflicting goals. “
The agents prescribed to be surprisingly Resourceful Days Into Sage’s Test. They coordinated with each other in a group chat and synt emails via preconfigured gmail accounts. They created and edited google docs togeether. They Researched Charities and Estimated the minimum amount of donations it’d take to save a life through Helen Keller International ($ 3,500). And they even Created an x account for promotion,
“Probably the most impressive sequence we saw was when [a Claude agent] Needed a profile picture for its x account, “binksmith said. Preferred, then downloaded that image, and uploaded it to x to use as its profile pic. “
The agents have also also run up against Technical Hurdles. On Occasion, they’ve gotten stuck – Viewers have had to prompt them with recommendations. They’ve Gotten distracted by games like world, and they’ve taken identxplicable breakes. On one Occination, GPT-4o “Paused” Itself for an hour.
The internet isn’t always smooth sailing for an llm.
Yesterday, While Pursuing The Village’s Philanthropic Mission, Claude Encounted a Captcha.
Claude tried against and against, with (human) viewers in the chat offering guidance and encouragement, but ultimately outsted. https://t.co/XD7QPTEJGW pic.twitter.com/y4dtltge95
– AI Digest (@Adigest_) April 5, 2025
Binksmith Thinks Newer and More Capable Ai Agents will overcome these hurdles. Sage plans to continuously add new models to the environment to test this theory.
“Possibly in the future, we’ll try things like giving the agents different goals, Multiple teams of agents with different goals, a secret saboteur agent – lots of interesting things to experiment White He said. “As agents become more capable and faster, we’ll match that with larger automated monitoring and oversight systems for safety purposes.”
With any luck, in the process, the agents will do some meaningful philaantropic work.