Learning Library

← Back to Library

Unified Risk Operations Center Strategy

Key Points

  • Cyber criminals exploit the fragmented, siloed nature of traditional risk functions—anti‑fraud, AML, SOC, insider‑threat, etc.—which leads to duplicated tools, data, and processes and creates gaps they can abuse.
  • A realistic attack (phishing → credential theft → SIM‑swap → crypto laundering) demonstrates how no single department has full visibility, causing each to misinterpret the incident and respond inadequately.
  • Building a Unified Risk Operations Center (UROC) consolidates data from every source and format across the organization, breaking down silos and providing a single pane of glass for risk insight.
  • The UROC relies on machine‑learning‑driven risk profiling and standardized, automated workflows that translate disparate data into consistent, actionable alerts.
  • By unifying data, analytics, and response processes, the organization can detect attacks that would have gone unnoticed, close exploitation gaps, and respond more quickly and accurately to threats.

Full Transcript

# Unified Risk Operations Center Strategy **Source:** [https://www.youtube.com/watch?v=bOPhZiW0-Rs](https://www.youtube.com/watch?v=bOPhZiW0-Rs) **Duration:** 00:07:59 ## Summary - Cyber criminals exploit the fragmented, siloed nature of traditional risk functions—anti‑fraud, AML, SOC, insider‑threat, etc.—which leads to duplicated tools, data, and processes and creates gaps they can abuse. - A realistic attack (phishing → credential theft → SIM‑swap → crypto laundering) demonstrates how no single department has full visibility, causing each to misinterpret the incident and respond inadequately. - Building a Unified Risk Operations Center (UROC) consolidates data from every source and format across the organization, breaking down silos and providing a single pane of glass for risk insight. - The UROC relies on machine‑learning‑driven risk profiling and standardized, automated workflows that translate disparate data into consistent, actionable alerts. - By unifying data, analytics, and response processes, the organization can detect attacks that would have gone unnoticed, close exploitation gaps, and respond more quickly and accurately to threats. ## Sections - [00:00:00](https://www.youtube.com/watch?v=bOPhZiW0-Rs&t=0s) **Unified Risk Operations Center Strategy** - The speaker argues that fragmented anti‑fraud, AML, and security teams create exploitable gaps, and proposes a machine‑learning‑driven, open‑platform unified risk operations center to detect and prevent attacks like phishing and SIM‑swap fraud. ## Full Transcript
0:00cyber criminals are becoming 0:01increasingly aggressive and they're 0:03counting on your organization to have a 0:05fractured response to their bad behavior 0:08but you can take back the upper hand by 0:10developing a new unified risk operations 0:12center strategy 0:14this strategy uses machine learning and 0:16open platforms to enable you to detect 0:19attacks that previously would have gone 0:21unnoticed 0:23so the problem with what we're doing 0:25today is risk management is really being 0:27built in silos we have a department for 0:30anti-fraud we'll have a different 0:32department for anti-money laundering 0:34we've got the security operations center 0:36corporate security insider threat and 0:38the the issue is that all of these 0:40different groups are duplicating their 0:42effort they have duplicate data tools 0:45and tactics and duplicate processes 0:48and ultimately this results in gaps and 0:50inconsistencies that cyber criminals 0:52really love to exploit 0:54so let's take a little look at an 0:56example scenario today we're going to 0:58have our attacker who's going after our 1:01unsuspecting victim jim 1:03so our attacker is going to send jim an 1:05email a phishing email which contains 1:07malware 1:09so jim's a nice guy he's not expecting 1:11to be phished today he opens that email 1:13which results in his banking credentials 1:15being sent to our attacker 1:18our attacker then uses these credentials 1:20to log into jim's bank account but jim 1:23was smart jim set up second factor 1:25authentication 1:26unfortunately our attacker is pretty 1:28crafty and steals jim's mobile number by 1:31performing an illegitimate sim swap 1:34so now he's able to approve the second 1:36factor authentication challenge which 1:38gives him full access to jim's bank 1:40account 1:42with this access he then buys 500 000 1:44worth of cryptocurrency which he then 1:47transfers to different mule accounts 1:49uses a crypto laundering service to make 1:52this unable to be traced back to him 1:55so we can have a think about how our 1:56different departments might detect and 1:58respond to this today 2:00our security operations center might say 2:02that this is actually just legitimate 2:04behavior there's nothing suspicious 2:06here our anti-fraud department might say 2:10that this looks like a case of 2:11compromised credentials 2:13and our anti-money laundering department 2:16might actually say that jim was the one 2:17who was laundering money 2:20so the point is all these different 2:21departments don't have full visibility 2:23into what took place which means that 2:25none of them can respond accurately and 2:28timely to the incident 2:31so how are we going to bring power back 2:32to the organization 2:34well we can bridge the gap by building a 2:36unified risk operation center strategy 2:39so this strategy uses uh makes use of a 2:43couple of core principles so one being 2:45using data wherever it lives in whatever 2:48format from across the organization 2:50consolidating that data to make it 2:51accessible 2:53we also use machine learning and risk 2:55profiling at the core 2:57and we build consistent workflows on top 2:59of this data and the machine learning 3:01insights that we get so that everything 3:03is handled consistently across the 3:05organization 3:06and finally this is a consumable service 3:09so that other parts of the business 3:10other applications and services are able 3:13to use the insights that we're getting 3:14out of our risk analytics 3:18okay so how do we get there this is an 3:20iterative journey it doesn't happen 3:21overnight what we do is we pick some use 3:24cases and we build on them each time 3:26proving value to the organization of 3:28doing so 3:29so to start with here we're going to 3:31start by integrating our anti-fraud and 3:34our anti-money laundering departments 3:36so at this point we've got a small swarm 3:38team which has got a couple of people 3:40from the different groups 3:42we're starting to do joint strategy and 3:44operations planning between these groups 3:46and we're also rationalizing the tools 3:48and controls between them 3:51so what can we detect now well our 3:53anti-money laundering department might 3:55have noticed that this money went to a 3:56crypto laundering service 3:58and our anti-fraud department might have 4:00noticed that you know this transaction 4:02took place at an unusual time of day 4:06so from here we can start to remediate 4:08we can start to take action on these 4:10insights so we could automatically block 4:13future access to this account 4:15and we can automatically notify the 4:17authorities so that we're preventing 4:18future losses to the business 4:21okay so our next step is then to scale 4:23up so now we're going to pull in 4:25information from our security operations 4:27center and our corporate security groups 4:31so now our swarm team is getting bigger 4:33we're including people from these new 4:35departments we're also further 4:37rationalizing the tools and controls so 4:39we're reducing that duplicate 4:41set of tools in the organization 4:44and at this phase of maturity you're 4:45really starting to make these insights 4:47consumable so other parts of the 4:48business can use the the rich insights 4:51that we're building here 4:53okay so what are we going to be able to 4:55detect now well the sock might be able 4:57to tell us that our organization is 4:59being targeted at the moment by a 5:01phishing campaign 5:03we might also know that jim had malware 5:05on his device at the time of compromise 5:08and we can start to pull in information 5:10from other third parties so the telco 5:12might be able to tell us that you know 5:14the sim swap that took place was 5:16actually illegitimate or in fact we 5:19could look up information about recent 5:20activity on jim's account 5:23so now what can we do with this 5:24information well we can see how the 5:26attacker got access in the first place 5:28which means we can start to remediate 5:30and ensure that this type of attack 5:32doesn't take place again in the future 5:35okay so now we want to do our final 5:37iteration we want our fully fledged 5:38unified risk operations center 5:41and so what we're going to do is we're 5:42going to make sure that all of the 5:44departments in our organization are 5:46integrated into this platform so we're 5:48pulling data from wherever it lives in 5:50the organization 5:52and what if whatever format it lives in 5:55we are able to do to consolidate this 5:57data so that we can do machine learning 5:59on top of it and we're getting those 6:00risk insights associated with all 6:02different types of entities throughout 6:04the environment we also have 6:06well-defined workflows so we can 6:08automatically deal with known threats 6:10and we have a really well-structured way 6:12for dealing with unknown threats 6:15and of course as i mentioned before this 6:16is all consumable so other parts of the 6:18business can use this information 6:21okay so now what can we detect at this 6:23final phase well all the pieces of the 6:25puzzle are starting to come together we 6:28might be able to see things like this 6:29device that was used to authenticate 6:33the attacker was actually a new device 6:35which in itself isn't an indicator of 6:37fraud but it does increase the 6:39suspiciousness of this particular 6:41incident 6:43we can also correlate this with other 6:44information from consortium data for 6:46example and we could say okay well we've 6:49actually seen this device be used for 6:51fraud in the past 6:54we can do machine learning on different 6:56parts of our environment so we could do 6:57machine learning on jim's usual 6:59transaction so the transaction value and 7:02the transaction type which would flag 7:04this particular transaction as an 7:06anomaly 7:07and ultimately what we're doing is we're 7:09pulling all these little pieces of risk 7:11information together to be able to 7:13inform real-time decisions so in this 7:16case we can prevent this transaction 7:18from ever actually having taken place 7:20anyway 7:23okay so we know that our existing 7:25mechanisms of building risk management 7:27around specific silos isn't sustainable 7:30we need to bring together the data the 7:32tools and the people in order to be able 7:35to effectively manage risk 7:37consider your risk management 7:39modernization journey and have a think 7:41about how unified risk operation center 7:43strategy could help you manage risk now 7:46and into the future 7:48thank you if you like this video and 7:50want to see more like it please like and 7:53subscribe 7:54if you have questions please drop them 7:56in the comments below