BAE Systems

 
In a recent webinar with UK Finance, Nordea and Credit Agricole, we touched on a number of issues regarding the financial sector information sharing landscape. In his latest blog, David Nicholson recaps the webinar discussion and covers the key technical and operational barriers, the benefits and a flavour of the technology required

 
Thursday 10 June
Read time: 3 mins
Information sharing between public and private organisations is a key enabler in the defence against financial crime. It enables information fusion to achieve better coverage and detection of the criminal behaviour. In the ‘Information fusion in financial crime’ webinar with UK Finance, Nordea, and Credit Agricole, we touched on a number of issues regarding the information sharing landscape. This included some of the key technical and operational barriers in this context, the benefits of information fusion and a flavour of the innovation and technology required to achieve these.
 
With a multitude of public and private organisations working (either directly or indirectly) within the money laundering and fraud arena, the UK information sharing landscape has become very complex. While a lot of relevant information exists within the ecosystem, it can remain stagnant within sectors and the information flows lack formalisation. The overarching challenge is to enable this information to be fully exploited for intelligence purposes by breaking down silos to facilitate the sharing of information between organisations and sectors as “business as usual”.
 
 

Incremental advances = efficiency gains


In the spirit of incremental advance, there is likely to be significant efficiency gains resulting in the first instance from breaking silos and joining the dots between different compliance functions within a bank. For example, by linking know your customer (KYC) and due diligence with transaction monitoring, compliance solutions can move from periodic reviews toward event driven reviews. This kind of perpetual or dynamic KYC, coupled with digital on boarding of customers and use of artificial intelligence (AI), is a clear step toward increased efficiency. 
 
A further incremental efficiency gain is possible when the compliance and fraud functions within banks pool their insights and intelligence, with yet further gains available when multiple banks join forces to create information sharing coalitions. A number of such initiatives are underway: for example, Transaction Monitoring Netherlands and Invidem, a shared KYC compliant service involving banks in the Nordics. These should provide a more complete, less fragmented, view of the transaction chain, in addition to providing further context to better track and detect the flow of laundered money.
 
The UK’s Joint Money Laundering Taskforce (JMLIT)  initiative, established in 2015, has demonstrated the value of public-private partnerships (PPP) through information sharing between law enforcement, justice agencies, government departments, regulatory bodies and the private sector. Similar initiatives have followed, such as the Swedish AML Initiative (SAMLIT).
 
 

Information fusion and information control


Enabling technology and innovation in support of exploiting information fall broadly into two categories: information fusion and information control. Information fusion technology enables the combination of formation from multiple PPP sources, providing a more complete, accurate, and intelligence-led view of the financial crime risk. Specifically, machine learning contains predictive models that expose weak patterns and anomalies in data. Federated machine learning is a framework that enables participating organisations to build a joint predictive model without sharing local data.
 
Given the complexity of the information sharing landscape, it is crucial to manage the system-wide flow of information. This is where information control is important. PPP systems are comprised of multiple stakeholders who need to take coordinated action to satisfy an overall ‘team’ objective, essentially a multi-agent system. Although not fully explored in the PPP context, multi-agent system technologies such as incentive engineering and coalition formation are likely to be important here. Technology such as provenance graphs may provide a sound framework to ensure and accountable information infrastructure to track and audit information flow and decisions within the PPP system.
 
 

Data privacy challenges


However, data privacy is perhaps the biggest challenge of all. Privacy-preserving technologies such as homomorphic encryption and the aforementioned federated machine learning are important potential enablers. Equally important, within trans-national PPP systems, are solid GDPR agreement frameworks for the purposes of anti-financial crime. It is also necessary to distinguish between operational and strategic information. Clearly, there will be different levels of privacy associated with operational data at the individual customer or entity level compared to strategic level data used at a more thematic level.
 
Managing the additional complexity imposed by PPPs presents a further significant challenge. In this regard (as noted earlier) a controlled and incremental approach to information sharing and fusion is key. Technology has an important role to play here by automating parts of the PPP system where there is less risk. This would allow skilled human resource to focus attention on the areas where risk is higher.
 
Arguably, an optimal PPP system is one in which information is freely shared by organisations to a central hub and the products of information fusion (operational and strategic intelligence) are distributed back to the organisations. However, full-scale centralisation is unrealistic. A more practical solution is likely to involve pockets of centralisation, e.g. between the KYC and transaction monitoring functions at a bank, and a more tightly controlled flow of information between banks, regulators and law enforcement agencies. Centralisation where possible is important, but what information is shared between participants and how it is processed is just as important.
 
Looking ahead, notwithstanding the undoubted challenges, the panellists expected to see increased effectiveness and efficiency in the fight against financial crime arising from growing levels of activity around PPP initiatives. However, effectiveness and efficiency will need careful definition in this context, not least so that stakeholders can provide quantitative evidence of the information fusion benefits to regulatory and supervisory bodies for securing confidence and buy-in. 
 

Other recommended content

Do you want to access more blog content from our Financial Services experts?

More blog content
banking sign up

 
Sign up to get the latest industry intelligence and insight

Stay on top of the latest news, forthcoming webinars, new podcast episodes and upcoming trends in Banking, Insurance, Data and Cyber by signing up to our BAE Systems Insights series. Hear from industry experts sharing their views on hot topics and new technologies. Delivered fresh to your inbox. 
 

 
 
 
Get in touch with our experts today

 

Americas  +1 720 696 9830     |     Europe, Middle East  +44 (0) 330 158 3627     |     AsiaPac  +61 290 539 330