What if every staff member had ten assistants?
What if every person on your team had ten assistants?
Right now most of us use AI to chat. We ask a question, get an answer and copy it somewhere useful. New tools like Claude Cowork and OpenClaw work differently. They take on a task and carry it out, and one person can run several of them at once. To me, that marks the end of the chatbot era and the start of the agentic era.
Imagine Canada’s recent report, The State of Artificial Intelligence Adoption in Canadian Nonprofits, shows 80% of organizations are using AI, but mostly for light tasks like drafting emails. That’s a long way from where these tools are heading.
From one chatbot to a team of assistants
My prediction is that by the end of this year, and likely sooner, security concerns will be a solved hurdle. Once they are, every worker will be able to spin up as many specialized AI assistants as they need, each handling a repetitive task at the same time as the others. Think third-party data cleaning, or research on individual donors.
The security work is already happening in real time. OpenClaw, for example, can run locally on your own computer, which helps keep your data private. I expect security to become a foundation you build on, instead of the barrier it is today.
Your job description is about to change
Your job title might stay the same this year, but your job description is going to change. As tools like OpenClaw let us orchestrate several AI models through a complex workflow, the most valuable skill for a nonprofit professional will be managing the assistants that do the work.
Take major donor research. Today someone does it by hand, or with a few one-off prompts. With a small squad of agents, one monitors the news, one analyzes giving history and one drafts a 12-month engagement plan. Your gift officer spends their time on the relationship.
Staff end up spending less time making things and more time directing and checking the work. That’s a real skill, and most teams haven’t been trained for it yet.
Teams stop waiting on each other
Most nonprofits run on linear workflows, where step A has to finish before step B can start. That creates bottlenecks and burnout. Comms waits for fundraising data, which waits for admin to process it.
With agents like Claude Cowork handling the shovelling work in the background, departments don’t have to wait in a queue. Your assistants can process, clean and summarize data across all three departments at once, and give leadership a real-time command centre for what’s happening.
How to get your team ready
- List the repetitive tasks. Ask each person for the jobs they do the same way every week. Those are the first candidates for an assistant.
- Write a brief for one assistant. Pick one task and write down what the assistant should do, what it can access and what a finished result looks like.
- Make review part of the job. Decide who checks the assistant’s work and how. Managing an assistant means reading what it produced before anything goes out the door.
- Lead with green flags. The report found 64% of nonprofits have no formal AI policy. That gap often gets filled with fear-based rules about what might go wrong. Give staff a list of safe, high-value uses too, like pointing a private AI tool at your staff handbook so new hires get instant answers.
- Add a second assistant once the first one works. Build up one at a time. Each one teaches your team more about how to brief an assistant and check its work.
For leaders: get your strategy ready
The report shows where most of us are: 59% of nonprofits use AI for marketing, and only 12% use it for staff development. We’re still working tactically.
Here’s how I see the difference:
- The tactical nonprofit, today. One-off prompts for emails, AI treated as a red-flag risk to manage, slow hand-offs between teams, and staff who make every piece of work themselves.
- The strategic nonprofit, where 2026 is heading. Teams of agents handling daily operations, AI treated as a green-flag way to multiply impact, real-time collaboration across teams, and staff who manage agents.
The technology is getting close. Before it gets here, I’d make sure your strategy and your team are ready for it. A few questions to start with:
- Which repetitive tasks would we hand to an assistant first?
- Who on our team is ready to manage AI work, and who needs training?
- Does our AI policy tell staff what they can do, as well as what they can’t?
If you want a starting point, my free Nonprofit AI Readiness Roadmap will show you where your organization stands today.