For years, digital transformation in banking was about moving services online, making transactions faster, automating back-end processes and improving the customer experience. But the next phase isn’t just about being digital. It’s about being AI-driven.
That shift isn’t just about chatbots handling customer queries or AI helping detect fraud. AI is being embedded into the core of financial services, shaping how decisions are made, how risks are managed and even how banks themselves operate.
Financial institutions are already seeing the impact. AI-powered systems are helping banks deliver hyper-personalised financial services, optimise fraud detection and improve risk assessments. AI-driven automation is reducing human error, making operations more efficient, and, in some cases, even transforming how banks interact with their customers.
Yet, with AI playing a bigger role in banking, an important question arises: Are banks ready for it? AI is placing unprecedented demands on financial infrastructure, requiring real-time data processing, seamless scalability and round-the-clock availability. And not all banks are equipped to keep up.
Tackling resilience
At Mobile World Congress (MWC) 2025, Huawei introduced the AI-powered R-A-A-S framework, a structured approach to making AI-infused banking resilient, secure and scalable.
The framework – short for Reliability, Availability, Autonomy and Security – helps financial institutions meet the operational demands of embedding AI into their systems. It builds on Huawei’s earlier ‘Four Zeros’ vision, announced at MWC 2024, which set ambitious targets around system uptime, automation and security: Zero Downtime, Zero Wait, Zero Trust and Zero Touch.
Alvin Feng, director of global marketing and solution sales at Huawei Digital Finance BU, says the shift towards AI-first finance means banks need a new kind of digital foundation. “AI is growing fast and bringing big opportunities,” he says, “but also major challenges – especially for infrastructure.”
That’s where the R-A-A-S framework comes in. According to Feng, its purpose is to give banks the resilience they need to keep pace with AI demands, particularly as new models become cheaper, smaller and more accessible. “AI is no longer only for big banks,” he says. “Smaller and mid-sized banks can also leverage these technologies.”
Alvin Feng, director of global marketing and solution sales, Huawei Digital Finance BU
Trust and transparency
Technologies like chain-of-thought reasoning, which make the steps of an AI decision visible, help meet regulatory demands and build trust. “AI isn’t just a tool for handling tasks,” he also said. “It’s becoming a trusted assistant in the core business of banking.”
That means AI isn’t staying on the edges of finance. It’s moving into the centre, helping with marketing, customer experience, decision-making and even the internal architecture of banking services. During MWC, Huawei pointed to a future where every bank has its own AI agent, trained on its own data, embedded across its operations. “There will be two types of employees,” said Feng. “AI employees and employees empowered by AI.”
Redefining the experience
This kind of vision isn’t hypothetical. Huawei demonstrated how, in China, AI agents embedded in mobile apps have reduced money transfers from 10 steps to just two. In risk management, AI systems are combining large models and knowledge graphs to spot fraud more quickly and with greater accuracy. In marketing, AI tools are helping account managers serve up to four times as many clients by identifying the right moment and method to engage.
It’s a model that Jason Cao, CEO of Huawei Digital Finance BU, sees gaining momentum outside China too. “Just in a few months, banks have started applying AI in new ways,” he said. “It’s already at a new level.”
Cao sees the most advanced institutions moving beyond pilot projects and beginning to explore how AI agents will reshape their core systems. “We think that the front, middle and back-office architecture may be totally disrupted,” he said. “The AI agent will be normal and play a significant role.”
Meeting the challenge
The R-A-A-S framework covers four areas:
Reliability, which includes real-time data synchronisation and multi-copy storage to ensure zero data loss
Availability, achieved through active-active data centres and cloud-based microservices to support 99.999 per cent uptime
Autonomy, with AIOps and automation to reduce human error and ensure that problems can be discovered within one minute, located within five, and resolved within 10
Security, using AI to detect and block threats within seconds and orchestrating protection across terminal, edge, cloud and network
Huawei at MWC 2025
A future that’s arriving fast
Cao believes the urgency around AI readiness is only going to grow. “More innovation means more pressure,” he also said. “We have to make sure the system can handle it.”
With customer behaviour shifting, expectations rising and the tools of AI becoming more accessible by the day, it’s no longer a question of when banks will adopt AI, it’s whether their systems can take the strain. For many, resilience could be the deciding factor between scaling AI successfully and falling behind.
The post AI is Taking Over Banking – But How Can Banks Keep Up? appeared first on The Fintech Times.
