Hi friends,
I have been out of the scene these past two months. At times it felt like I was racing against time, so I had to stop writing.
I had promised myself I'd keep to a biweekly cadence, but for the first time since starting this newsletter, life had other plans. I had to let go of the noise across several fronts, including AI noise, and focus on what really matters.
I am back now, and I hope to be more consistent from here.
So just to give you a short life update. This period yet again has been a reminder that life is unexpected and that we need to spend it more deliberately on what matters to us.
I spent much of the last two months in Northern Spain, where my mother received a cancer diagnosis on top of several existing health issues.
The weeks that followed were filled with hospital visits, specialist appointments, difficult decisions and trying to navigate a healthcare system in a language we don't speak. AI translation tools became unexpectedly indispensable as we worked through test results, treatment options and costs.
As a family, we eventually decided that the best place for her treatment would be back home in Kazakhstan, where I will now spend the coming months supporting her recovery.
It wasn't all difficult. We still managed to spend time by the sea, laugh together and make memories that I'll always treasure. Those moments reminded me how quickly professional priorities fade into the background when life suddenly changes.
Ironically, while I had switched off from AI in financial services, I found myself relying on AI almost every day—not for work, but to translate conversations with doctors, understand medical terminology and navigate an unfamiliar healthcare system. It reminded me that the most useful technology is often the technology you stop noticing. During those weeks, the whole AI conversation felt very far away under the hot Spanish sun, and perhaps that was exactly what I needed.
Regulatory updates
With that said, I haven't been following the industry nearly as closely over the past two months. Catching up this week, though, one theme immediately stood out.
There has been a move from the regulatory side, though. On 6 July, the FCA published its Mills Review, setting out how AI could reshape retail financial services by 2030, and on 10 June the Financial Stability Board published its first operational framework for governing agentic AI, built around board accountability and lifecycle management.
Both are asking the same ownership question I raised in my last issue. The difference is where the question is coming from. Mine was operational: what actually happens inside an institution when a workflow runs through an agent instead of a person. Theirs is supervisory: what a regulator expects to see, and checks for, from outside the institution.
If you do one thing before the next issue
This is a new recurring section. Each issue I will leave you with one specific insight, question, tool or case to sit with. For this one:
Pull up your own team's or organisation's AI use case list that you have visibility into, and ask one question against each: who owns the outcome if this goes wrong? That is the question sitting underneath both the FSB's June framework and the FCA's Mills Review, and it takes around five minutes to ask, even if it takes months to answer properly.
I will review both reports in more detail in my coming issues.
That's the wrap, and I am so glad to be back.
- Gaziza
P.S. On another note, I am looking to connect with people in my network who work in compliance, risk, operations, and specifically AI, based in the UK with an EU-facing remit. If that is you, please give me a shout. I would like to have a chat.
Thanks for reading. If this one landed for you, sharing it with someone who'd find it useful means a lot. |