AI Governance: Wild West For Banks As Their AI Run Wild
It's not if, but when some bank gets unlucky and an AI goes rogue.
This is my daily post. I write daily but send my newsletter to your email only on Sundays. Go HERE to see my past newsletters.
HAND-CURATED FOR YOU
This report on how banks and financial services (BFS) can become AI-first is good, but its survey of 505 leaders reveals that we’re still in the “Wild West” days of AI governance, and many are courting disaster.
This HFS Infosys report is a great read, and I want readers to know that I’m not picking on it. It is a great window into bank GenAI use.
That said, as with other recent extensive AI surveys, the window it opens shows a repeated pattern of banks falling far behind on their requirement to govern AI.
If you think I’m exaggerating, this report shows that just 23% of BFS enterprises had mature AI governance and risk management practices.
Readers may think that is excusable given how new AI is and that the keyword “mature” is perhaps asking too much. This may be true, as the survey showed that 65% of respondents had only a dedicated AI budget for less than 2 years.
Yet still, you’d think that any BFS running an AI would know what to do if the AI goes rogue, right? That would seem like nothing more than common sense. Something more formal than an “Off” switch.
So I’ll let you decide whether the “Wild West” label fits when 78% are not “fully matured” in “establishing incident response procedures.”
You can’t make this stuff up.
Banks are heading headfirst into the AI age, with GenAI systems that are acknowledged to hallucinate, and they are unprepared to handle incidents.
That’s why I’m calling this the “Wild West” days of AI, where instead of gunslingers, we have AI slingers who will do whatever they want until the law comes knocking.
Just like in any Wild West movie, the law will arrive.
👉Banks’ AI Governance Problem by the Numbers
🔹 AI Governance is a problem: only 23% are considered to have “fully matured” practices. This means that 77% are running AI without appropriate governance
🔹 But governance doesn’t matter, AI is trustworthy: Think so? Only 84% do not have mature programs to “integrate trustworthy AI characteristics.” How can anyone know for sure?
🔹 Staff are ready for emergencies: 78% do not have mature incident response procedures
🔹 But of course AI operations are transparent! Sorry only 73% are not mature in this area either!
🔹 Setting the AI strategy: 79% across various tech roles specified that IT sets the AI strategy.
🔹 Top 3 AI uses: Data analysis, Marketing and Process automation.
🔹 Top 3 AI challenges: Data quality and access, Security and Privacy, Talent
🔹 Top 3 objectives for AI investment: Productivity, Efficiency and Profitability
If you know someone who would like this newsletter, please share it with them and help grow our Asia, CBDC, and AI aficionados community!
