There are two sentences an arts leader can say about AI, and they lead to two different organisations.
The first: "AI will let us do what we do now with fewer resources." The second: "AI will let us do more of what we exist for, with the resources we have." Same technology. Same tools. Opposite institutions at the end of the road. If you only have bandwidth for one strategic thought about AI this year, we'd nominate this one, because every other decision — which tools, what policy, whose time — flows downstream of it.
Efficiency AI: the same with less
The first sentence is the default, because it's the language of every technology procurement of the last thirty years. And in a funding crisis it has real gravitational pull: if the minutes write themselves and the reports half-draft themselves, the spreadsheet-brained conclusion is that you need fewer hours of administrator — so the post shrinks, the saving is banked, and the output stays exactly where it was.
Notice what this does. The staff who remain absorb the checking, the anxiety and the message that the technology's purpose is subtraction — so they resist it, sensibly, because it has been aimed at them. The organisation gets no better at its mission; it becomes a slightly cheaper version of itself, run by people who trust it slightly less. And once funders can see the sector runs on fewer people, standstill funding quietly becomes the new generosity. Efficiency AI doesn't just shrink organisations; it shrinks the sector's case for support.
Opportunity AI: more with the same
The second sentence treats the recovered hours as a windfall — found capacity, already paid for — and spends them upward, on the mission. The administrator who no longer loses Mondays to minutes runs the participation programme's waiting list down. The marketer freed from reformatting listings finally does actual marketing. The producer's recovered evenings become one more funding application — or, radical thought, an evening off, in a sector that runs on burnout. The access work that was always deferred stops being deferred. Output grows. Reach grows. The people stay, and the technology is experienced as a gift rather than a threat — because it is one.
Making it real (because a slogan won't survive budget season)
- Declare it. A public sentence, in the AI policy and in front of staff: "Time saved by AI goes to mission and working conditions, not to headcount reduction." Written down, it's a commitment people can hold you to — which is what makes it worth something.
- Count the dividend. Track the saved hours, roughly — a shared note will do. What gets measured gets protected; invisible savings get silently absorbed.
- Spend it visibly. Name the things that happen because of the recovered time: the extra school workshops, the audited access pages, the fourth funding bid. When the board asks what AI did for the organisation, the answer should be a list of mission, not a percentage of cost.
- Hold the line in the hard year. The test arrives with the next funding cut, when "the same with less" comes knocking dressed as prudence. An organisation that has counted and named its dividend can show the cut what it would actually destroy. That's the argument working itself.
The technology industry will keep selling efficiency, because efficiency is what buyers of software are assumed to want. Arts organisations get to want something better: not a cheaper version of themselves, but a fuller one — more art, more people, more of the point. Same tools. Your sentence chooses.
The image above was AI-generated — as ever, we'll always tell you.