Every arts organisation has one: the System. The spreadsheet only Sandra understands. The membership database built in 2009 by a volunteer who moved to Portugal. The box office export that has to be retyped into the finance sheet every month because the two systems have never spoken and never will. So when leaders ask "can AI help us modernise our systems?", the yearning behind the question is real. The answer is yes — genuinely, usefully yes — but not in the way the sales emails suggest, and the difference matters.
What AI won't do
Let's clear the fantasy first. AI will not look at your organisation, select the right ticketing platform, migrate your data overnight and retire Sandra's spreadsheet with a farewell card. System change is mostly decisions — what you actually need, what you can afford, who does what — and decisions remain stubbornly your department. Anyone selling "AI-powered digital transformation" as a product is selling consultancy with the lights turned up.
What it will do is stranger and more useful
- It can finally explain your own systems to you. Paste in a sample of the mystery spreadsheet (strip the personal data first — house rules) and ask: what is the structure here, what's inconsistent, what would break if we changed it? Institutional knowledge that lived in one person's head becomes a written document. Do this before Sandra retires, not after.
- It can do the cleaning. Every modernisation project dies in the data-cleaning phase: eight hundred duplicate contacts, dates in three formats, "MISC2". AI writes the rules, the formulas and the step-by-step process for fixing this — the single most tedious and most necessary job in any migration.
- It can translate between you and the technology people. This one is quietly transformative. Ask AI to turn "we need the box office to talk to the mailing list" into a proper requirements list; ask it to explain a vendor's proposal back to you in plain English; ask it what questions a cautious buyer would ask about that quote. Organisations get fleeced at the moment of translation — walking in with a competent translator changes the negotiation.
- It can build the small things. The rota tool, the room-booking form, the report that cross-references two exports — the little conveniences that were never worth a developer's day rate. With AI coding tools (the Codex and Claude Code end of the family), someone slightly technical on your team can now build these in an afternoon. One honest caution: internal conveniences, yes; anything public-facing or holding personal data needs proper professional care — a home-made tool that loses participant details is a modernisation story with an unhappy ending.
The realistic path
Not a transformation programme. A sequence: understand what you have (AI documents it), clean what you'll keep (AI does the drudge work), fix the small frictions (AI builds the little tools), and only then decide the big things — with AI as your translator and your second opinion, and humans making the calls. Some organisations will discover at step four that they don't need the expensive new platform at all; they needed their existing mess understood and tidied. That's a slightly deflating conclusion to a modernisation journey, and it saves five figures, so deflate away.
The Systems were never really the problem; the problem was that understanding them cost more time than anyone had. That's the cost AI just collapsed. Sandra's spreadsheet is finally documentable, the vendor's jargon is finally translatable, and the modernisation conversation your organisation has been deferring for a decade is — suddenly, unglamorously — affordable to have.
The image above was AI-generated — as ever, we'll always tell you.