Let AI Identify the fraud
invisible to the human eye
Sight to identify Anomalous
Transactions in your data
Enhance your sight with AI
See in 3D
Enhance your sight with AI
An AI Named Mantis
Mantises are the only invertebrates known to see in 3D.
Our AI solution looks at the data in various dimensions, in ways that the human eye alone is not able to see. AI powered sight will aid you in identifying anomalous, or suspicious, transactions in your data, after analyzing large volumes (millions of records) of historical data.
What is Anomaly Detection?
Anomaly detection is a way to capture suspicious events, which differ significantly from the majority of the data. Detecting anomalies can help authomate auditing transactions for fraud, to support reclaim efforts and enable preventive actions going forward.
This technology is getting more and more popular as fraud remains a serious issue across industries and the traditional methods can not adapt to the changes over time.
* According to the National Health Care
* According to Forbes
* According to the FBI
AI is the new solution to these old problems.
Why do we need AI for Fraud Detection?
Humanly Not Possible
Data Volume and Complex Patterns
Due to data flood and the complexity of fraud patterns, humans provoke excessive false positives and overlook false negatives. Contextual and Collective anomalies patterns are also very difficult to notice with the human eye.
If even 10% of 10 million annual expenses are marked problematic, that means a million expense line items would need to be investigated by the compliance team annually.
Adapt to New Changes
While humans usually look for known patterns, AI machine learning can adapt to new changes over time, search for unknown patterns, and keep up with anomalies.
Mantis is a no-code software applying AI to automate
the fraud detection process to catch suspicious events
which differ significantly from the majority of the data.
Collective, Contextual, and Point anomalies will
automatically be distinguished and displayed
for review in intuitive real-time reports.
Mantis is a no-code software applying AI to automate the fraud detection process to catch suspicious events which differ significantly from the majority of the data. Collective, Contextual, and Point anomalies will automatically be distinguished and displayed for review in intuitive real-time reports.
AI with ROI:
Yields immediate tangible financial benefits
Enable Upstream Audits at the Time of Estimate
Reduce systemic, recurring leakage by catching fraud upstream (during time of estimate) to catch fraud before invoiced.
Aid in Reclaim process post invoicing
Provide insights with interpretable reports, including benefit summary and visualization to guide user through further investigation and improve efficiency of audit reclaim process.
Leverage state of art machine learning models using a simple, guided user interface. This solution abstraction away from coding, enables Citizen Data Scientists in organizations to automate audits with machine learning.
Mantis flags anomaly that is invisible to human eyes due to the data flood and complex patterns. Stop recurring fraudulent transactions at a batch level and prevent blind spots, guiding users with interpretable, quantified results to further investigation.
Fraud is usually committed by multiple actors and factors at the same time, and we are not only able to catch a single fraudulent actor, but also the collusion between actors, such as repair billing fraud, inventory return abuse, non-rendered services, duplicate charges, collusion of doctor-patient/doctor pharmacies, fake shops/vendors conspiracy, in addition to identifying anomalous patterns for an actor or transaction level.
Utilize different algorithms based on various scenarios and assumptions to catch anomaly in different dimensions.
Citizen Data Scientists:
Whether your business has an audit team or not, Mantis enables users to leverage a wizard driven machine model configurator to guide you in building and visualizing anomalies in a consumable fashion to reduce errors. Without hiring additional data scientists, you can easily leverage AI to continuously improve internal preventive controls, mitigate regulatory risks, enhance the visibility of information, the augment quality of decision making processes, and create a culture of compliance.
In this human machine collaboration, humans still play significant roles in the process. The solution we offer is to extend from technology to business, a life cycle that starts from identifying anomalies via the output of AI models, and ends with benefit realization.
Employees can easily change the underlying assumption based on different business scenarios and needs after building the model.
The interpretable real-time reports with visualization generated by AI will guide users through the investigation process, empowering employees with unrevealed insights and taking further actions.
Leverage AI to enhance the quality of process and decision making, stop revenue leakage, and realize benefits.
Billing Invoices Fraud
As the global supply chain complexity increases, so does the flow of goods, money, and the information among business entities. Providing visibility to upstream transactions that involve suppliers and logistic providers in a more efficient manner, and managing them in a timely and automated way, has become a challenge for every company. Enhancing the visibility of the flow of goods, money, and the information between business entities, from the shipment to the transaction receipts, is the first step to prevent any potential disruption.
Our AI solution is able to flag fraudulent transactions and conspiracy of fraudulent actors, such as asset repair bill fraud, inventory return abuse, fake shops/vendors collusion, and duplicate charges, efficiently and automatically.
Pharma & Healthcare Billing Fraud, Waste & Abuse
According to the National Health Care Anti-Fraud Association, 3%-10% health care spending is fraudulent (up to $300B). In fiscal 2019, the Department of Justice recovered more than US$2.6 billion claims relating to the healthcare industry out of US$3 billion from civil cases involving fraud and false claims against the government. There is no precise measure of healthcare fraud waste and abuse, as we will never really know how many we missed out via false negatives, or accidently classifying fraud and not fraudulent.
Our AI solutions will highlight fraudulent transactions and actors, as well as the collusion of multiple fraudulent actors, such as doctor-pharmacists and doctors-patients conspiracy, instead of simply catching systemic fraud patterns at a batch transaction level and flagging a single actor.
Anti-Money Laundering (AML)
Banks in the U.S. spend more than US $25 billion annually, on average, on anti-money laundering compliance according to Forbes. As the volumes, complexity, availability, and regulations change.
AI will adapt to new changes over time and flag recurring fraudulent actors to augment the traditional rule-based monitor systems. Instead of giving binary feedback simply based on a threshold, which may be irrelevant overtime, AI will know the likelihood of fraudulent events. Compared to other machine learning solutions in the market, the active learning AI methodology we leverage detect the anomaly patterns and flag fraudulent actor collusion with high accuracy.
The total cost of insurance fraud, excluding health insurance, is estimated to surpass $40 billion annually, according to the FBI. Insurance Research Council (IRC) also estimated that up to $7.7 billion of auto injury claims was excessive payments in 2012, accounting for 13%-17% of the total payments of 5 main private passenger auto injury coverages.
Traditional anomaly detection technology in the insurance industry identifies anomalies by fitting them into a preprogrammed template. As transactions get more and more complicated, so does the fraud. A proactive approach is needed to adapt to new fraud patterns and perform dynamic analysis. AI is the only feasible solution that can keep up with anomaly detection and flag recurring fraudulent actors and transactions, when fraud is continuously adapting over time.
Supply Chain Sourcing, Production, & Quality Defects
The more components and processes in the production line, the greater the remake time and costs. For example, if we require 50% yield rate of the final product, we only need 50% of yield rate of the work-in-progress, if there is only 1 part needed to finish the production. However, if we need 10 parts to assemble the final product, then 93.3% of yield rate is required for each part, to maintain the 50% yield rate for the final product. However, 50% yield is unacceptable to most companies. Revealing the root cause of defects in each manufacturing process to enhance the yield is one of the most inexpensive ways to guarantee the quality, shorten the lead-time, and stabilize the supply chain.
AI can easily detect whether the outliers derive from the nature of process, human factors, machine factors, time factors, or environment factors, enabling stressors to be eliminated or controlled prior to any significant financial impact, and help you to identify improvement opportunities to achieve Six Sigma (3.4 defects per million, 99.99966%). This technology can also be applied to assist in making sourcing decisions. If there is a significant fluctuation in certain material prices in particular time or regions, maybe it is time for the sourcing department to investigate the reason and make a plan B.
According to McKinsey & Company’s estimate, the U.S. government suffers a US$150 billion dollars loss due to potential fraud annually, half of which goes undetected, and our AI anomaly detection solution enables the government to stop the revenue leakage in many ways. For example, the algorithms can easily identify the concentration of payments based on the geographic and demographic data the government has, thereby capturing the suspicious groups in suspicious regions. For another example, the government can utilize the technology to monitor any significant changes in the financial behaviors/status of tax and debt payers to forecast the possibility of insolvency. Due to data flood and the complexity of anomaly patterns, it is inefficient for humans to reveal these patterns and determine the hidden links between fraudulent actors.