Use of the model in financing
Risk management
From risk management, there was no shortage of lessons from the global financial crisis.
One of the greatest privileges of my career was to be posted from Morgan Stanley to work with US Finance Minister Hank Paulson and many of you at the Federal Reserve for the painful period. I will never forget: After working with the US Treasury and Federal Reserve at Fannie Mae and Freddie Mac, I went home Sunday after the Lehman competition – only to get a call to get back to new bold. We learned that AIG would be out of liquidity by Wednesday due to counterparty risk in a derivative subsidiary of the UK.
It led me to the very important lesson that I have repeated ever since: You can’t drive a car with mud on the windshield. You can’t run any institution with mud on the windshield. You need to clear it away – which is what data and risk analysis provides.
AI and ML can be a strong help to “clear the windshield.”
Consider anti-whitewashing money. Instead of a single marked transaction that triggers a manual study that consumes hours or days to analyze, AI patterns look over the entire network in real time. With AI you can see a newly created business account with dozens of small deposits from all over the country – each just under $ 10,000 reporting threshold. It then crosses the business director’s name against global sanctions and media reports and marks a hidden link to a high risk unit. Then the security team warns before The funds are wired offshore. One of Google’s bank customers experienced three times more financial crime risk, 60% fewer false positives and a 50% faster path from detection to action.
Another critical example of early detection and risk management is about cyber security. The bad news is that attacks on financial institutions – already a primary goal – are increasing. The good news is that the tools are available to counteract these threats.
At Google, CyberSecurity Sacrosanct – is embedded deep in our culture. Google was built from scratch with an intense paranoia about anyone who violates personal data, and with a mania -foal focus around any element of fortification, including what is called “Zero Trust” that requires strict identity verification and authorization for any access request, whether the user or device is inside or outside the network. This is what we use with our business partners – public and private sector.
Still, we know that we have to continue to increase our games because the evil is doing. Therefore, we continue to invest massively in cyber security. That’s why we acquired Mandiant. Mandiant’s insight emphasizes the need for a layered security method-robust multifactor approval, constant patching and better internal detection features. Right now, the global median is “residence time” – which means the time to discover an incident – is 11 days. The faster an incident is detected, the less injury an attacker can inflict.
That’s also why we utilize Google Deepmind – Google’s AI Research Laboratory. Recently, we had a breakthrough in applying AI to security threats showing what is possible: Google Deepmind introduced an AI agent called Big Sleep. This agent was developed to proactively hunt for unknown software vulneresses. We were glad to see it finding its first mistake in the real world in November, proving that AI can connect security holes before they are exploited.
Since then, it has revealed several critical vulnerabilities, including a recent one known only to threat actors. Using Intel from Google Threat Intelligence, Big Sleep predicted that the mistake was about to become the weapon – and shut it down. This is believed to be the first time AI directly stopped a living exploitation in nature. Google is now using this breakthrough beyond Google and helps secure open source software over the Internet.
Google will continue to push the border because this threat is growing and in -depth – both financially and reputation. I am convinced that AI must be used to keep up with the growing threat horizon.
Operational efficiency and efficiency
In operational efficiency and efficiency I would call three examples:
First, AI is seen on customer support increasingly as the gift that continues to give. It provides operational leverage and clears the most basic questions, giving customer support professionals the opportunity to focus on the more complicated problems. More important, historically, when a call came in customer support, it did not catch the questions of the queries and analyzed the root cause. Here, anyone who implements an AI contact center is actually able to gain insight from the customer center. I was recently with a CEO who said that the company first rolled AI in customer support, the employees indicated that they wanted more of it. They said essentially, “This has removed that drudgery. Our team has more time to think.”
Secondly, one of the most powerful new tools is something Google built called Notebooklm. With notebooklm you can load reports, articles, videos, audio files in it – and it will ingest and analyze. You can ask it to spot trends or pick out details. You can make it synthesize results and deliver them to you in the form of a podcast that you can interrupt with questions.
Third, organizations see an extraordinary boost in the developer’s efficiency and productivity with AI and tools like Coding Assist, which is something Google and many others offer. We hear this repeatedly from banks as a priority.
Innovation and growth
The last category: AI can support and accelerate growth options – to add alpha.
This innovation can be in the tech stack itself, by which I mean chips, models and applications. We are talking to a number of banks and other companies with financial services on the use of TPUs within their system to improve their trade operations.
In the applications, it can come in the solutions that allow organizations to get closer to customers – especially with Agentic AI. An example that my colleagues are excited about is the opportunity to tackle so -called Henrys: high revenue, not rich yet. The top financial institutions employ tens of thousands of financial advisers who dedicate their time to know, understand and serve clients with high net worth. However, the white glove model cannot scale to tackle the fastest growing wealth segment – the mass wealthy and “Henrys.” This means that there is a “council gap” and a miss the opportunity to build relationships with the next generation of top clients.
This is a huge opportunity to implement Agentic AI systems that combine the intelligence of advanced AI models with access to tools so that they can hoover in information and understand, reason and act across complex workflows – and take actions on your behalf and under your control. This allows advisers to gather hyper-personalized recommendations for clients, reducing human preparation time by over 60%, which in turn frees advisers to focus on building relationships.