Marketing teams rarely get to do one thing at a time. A campaign is in production, the social content needs adapting, a presentation needs updating, and another brief has just landed.
AI tools can take some of that load. The harder question is whether they make the work better, or just quicker.
In financial services, that question matters more than most. AI is already common across the sector: in a 2024 survey by the Bank of England and the FCA, 75% of the 118 firms that responded said they were already using AI, up from 58% in 2022. But using AI and using it well are not the same thing.
Speed is the easier gain
The quickest wins come from the repetitive parts of the process. We have found AI useful for:
summarising long briefs and background documents
organising research before a project starts
exploring early creative routes
adapting existing content into different formats and lengths
None of this sounds dramatic on its own. Across a busy marketing calendar, it adds up.
Speed matters most when work arrives at short notice. A leading UK financial services institution was preparing for its global investor event, and its in-house team needed extra support from a partner who understood financial services compliance. We joined as an extension of the team and produced five films on a short turnaround, all delivered ahead of the event.
AI wasn't the point of that project. But the principle is the same: when the pressure goes up, you need ways to move faster without lowering the standard.
Quality still depends on judgement
AI can give you ten headline options in seconds. Deciding whether any of them is right is the hard part.
A headline can be grammatically sound and still fall flat. A visual can look polished and say very little. Copy can answer the brief without sounding like the brand.
AI works best when it supports judgement, not when it replaces it. It can widen the options. People still have to decide which idea has potential, which line needs rewriting and which route should be dropped.
What you put in also shapes what you get out. A vague brief usually leads to vague work. The clearer you are about the brand, the audience, the objective and the constraints, the more useful the output becomes.
That's one reason an extended team that already knows your business gets more from these tools than one starting cold.
The rules don't change because AI wrote the first draft
There are no AI-specific advertising rules in the UK. The existing ones apply in full, and the responsibility stays with you.
Advertising: in June 2026, the ASA reminded advertisers that the CAP Code applies to AI-generated content in the same way as anything else, and that the advertiser is responsible for any AI output used in its ads.
Labelling: CAP's guidance says there is no blanket legal requirement in the UK to disclose that AI was used in an ad. But a label won't rescue an ad that is misleading in the first place.
Financial promotions: the FCA oversees AI through its existing, technology-neutral rules, including the Consumer Duty and the Senior Managers and Certification Regime. Communications with customers still need to be clear, fair and not misleading.
Industry guidance: the IPA and ISBA published 12 principles for using generative AI in advertising in 2023. The Advertising Association followed in 2026 with a voluntary best practice guide that builds on them.
In practice, a faster first draft still needs the same checks: accuracy, claims, product details, tone and compliance sign-off. Someone has to take responsibility for the final version, and it can't be the tool.
Four questions to ask before you use AI on your marketing
Whether it's your own team or an agency using AI on your behalf, ask:
Which tasks will AI help with, and which stay with people?
Who checks every claim, figure and product detail before it goes out?
What information goes into AI tools, and are we allowed to share it there?
Who signs off the final version?
So, does AI improve both?
For speed, often yes. For quality, only when it's used well.
We see AI as a way to cut repetitive work, explore more routes and make production more efficient. The quality still comes from context, experience, judgement and knowing when the first answer isn't good enough.
That's where AI fits best: not as a shortcut around good people, but as a way to give them more time for the work that needs them most.
