As artificial intelligence becomes increasingly embedded in the workplace, Anthropic’s AI fluency chief says one of the most important skills is knowing when not to use AI.
Swanson, who leads AI fluency research and learning at Anthropic, says a significant part of her role involves helping people understand the limits of AI—not simply teaching them how to use it.
“I have a job teaching people how to use AI, and I probably spend at least half of my time telling people when not to use AI.”
Knowing When Humans Should Do the Work
The comment highlights a growing challenge for workers adopting generative AI: determining which tasks should be delegated to AI and which require human involvement.
AI tools can help people brainstorm, summarize information and complete routine work, but relying on them for every task can potentially remove opportunities for people to develop skills, exercise judgment or engage directly with a problem.
For Swanson, AI fluency therefore goes beyond knowing which tools or prompts to use. It also involves recognizing situations where doing the work yourself may be the better choice.
AI Fluency Is More Than Prompting
The idea reflects a broader shift in how companies are approaching AI education. As AI becomes more capable, employees increasingly need to understand not only how to work with AI, but also when AI adds value—and when it doesn’t.
For Anthropic, that distinction is an important part of building effective AI users. The goal isn’t necessarily to automate every possible task, but to help people make better decisions about where technology fits into their work.
Swanson’s message is straightforward: being good at AI doesn’t mean using it for everything. Sometimes, the most effective AI decision is knowing when to do the work yourself.
