Mind Map Gallery AI in Insurance
- 1.0k
- 16
- 13
AI in Insurance
This comprehensive mind map tells us how AI technologies affect the insurance industry.
Edited at 2021-07-31 11:33:41
- Recommended to you
- Outline
AI in Insurance
Robotics
Physical bots
Drones
Physical aerial survey post natural catastrophes
Rogue drones
Remote assessment of disasters for property claims
Autonomous vehicles
Self driving cars
Ethically incorrect decision making
XXXX
Cognitive bots
Hyperaautomation
Auto payments & collections
Data leakage
Automation of mundane tasks
Task estimation & automation
Fraud
High operational efficiency
Customer Services
Chatbots
Voicebots
Voice recognition
Fake voice / voice cloning
Storage & security of personal data
Impersonation
No manual ID&V required
Upsell / cross sell
Phishing
Data based product identification
Advice
Incorrect product recommendation
Financial decisions not in interest of the clients
Data based personalised advice
Videobots
Digital twin
Deep fakes
ID theft
Storage & security of personal data
Empathy, virtual interaction
Behavioural biometrics analysis
Digital fingerprints
Data theft
Storage of biometric data
Data theft
Storage & security of personal data
Authentication through biometrics validation
Monitored key strokes, swipe & touch dynamics, mouse behaviour, finger swiping patterns
Snooping keystroke latency mimicking user behaviour
Storage of biometric data
Robot based attacks to mimic user behaviour
Detection of online potential fraudulent behaviour
Storage & security of personal data
Connected worlds
Smart home connectivity for property insurance
Sensors for water & electricity leakage
Data theft
Inflated data readings
Claims prevention through real time alerts
Fire & smoke detectors
Sensor misbehaviour
Claims prevention through real time alerts
Keyless entry
Access to personal home data
Take over of home & entry access
Access to trusted repair network providers
Storage & security of personal data
CCTV cameras
Intrusive data capture
Video footage for theft & burglary enabling claims approval
Access to personal home data
Storage & security of personal data
Telematics for motor insurance
Collision management systems
Access to personal home data
Remote access to vehicle
Take over of critical safety features
Storage & security of personal data
Driver behaviour analytics
Auto claims alerts to insurer
Usage based insurance
Vehicle usage patterns
Wearables for Healthcare
Smart watches & fitness trackers
Automatic gathering of personal info
Live location tracking
Storage & security of personal data
PI data theft
Enhanced personalised underwriting of premiums
Defining of new customer segments
Body sensors
Access to personal data
Intentional incorrect data provision
Remote sensor corruption
Storage & security of personal data
Real time health monitoring & alerts to hospitals, care providers & insurers
AI based process terrain management
Quality management
Quality assurance
Real time feedback to workforce
Customer feedback
Sampling
Bias / influenced sampling
Random sampling
Results & analytics
Inflated negative feedback (volumes & scores)
Non person dependent
Digital process mining
Process mapping
Theft of company proprietary data
Non person dependent
Computer vision
Data theft
Storage & security of company data
Process articulation & analysis
Monitored key strokes, website tracking
Snooping keystroke latency mimicking user behaviour
Invasion of employee privacy
Robot based attacks to mimic user behaviour
Storage & security of company data
Process variance analysis
AI based analytical models
Data analytics
Fraud analytics
Fraudulent invoicing by supply chain management providers
New sources of fraud not fed into the models impacting efficacy
Prediction & identification of potential fraud
Liability prediction & decisioning
Incorrect decisions introduced
Enhanced liability management
Reserves prediction
Incorrect predictions introduced
Inflated reserves introduced impacting forecasting, top line and P&L
Loss estimation
Enhanced management of claims losses
Inflated losses introduced impacting company P&L
Litigation propensity
Biases introduced
Discriminatory litigation
Image analytics
Image recognition through computer vision
Biases introduced
Embedding of malware
Auto identification of accident images
Real time estimation of repair costs
Video analytics
Video recognition
Embedding of malware
Auto liabiility decisioning through video analysis
Estimation of repair costs
Rules based decisioning engines
AI assisted decisioning
Underwriting & pricing modelling
Discriminatory models
Data & ID theft
Storage & security of company / personal data
Multi parameter approach
Adoption of macro influencers on models
Structured rules based underwriting models
Personalisation of policy creation
Real time pay-as-you-go models
Complaints management
Introduction of biased / discriminatory results
Data & ID theft
Storage & security of company / personal data
Multi parameter approach
Rule set based approach
Standardised & consistent output
AI based document automation
Cognitive OCR
Data extraction & validation
Introduction of malware in image content
Storage & security of personal data
Non person dependent
Data enrichment
Introduction of garbage data
Storage & security of personal data
Adoption of steganography
Redaction of text in images
Non person dependent
Incorrect redaction of text exposing sensitive information
Document creation / content management system
Data capture, format & transformation
Unauthorised & secure content embedded & distributed
Adoption of steganography
Auto generation of documents
AI in health insurance
Integration of healthcare & insurers
Symptom checker
Inclusion of corrupt data leading to distortion of end result
Patient data storage & security
Increased access to healthcare
Hospital risk & claim prevention
Real time alerts for patient trouble
Patient experience
Fraudulent diagnosis
Patient data storage & security
Straight through processing (cashless treatment
Personalised health care plans
Admin tasks
Fraudulent billing & invoicing
Patient data storage & security
Integrated systems = no paperwork
Diagnoses
Medical imaging
Lack of explicit labelling of algorithms
Black boxing of algorithms
Introduction of algorithmic bias
Inclusion of corrupt data leading to distortion of end result
Patient data storage & security
Recognition of complex patterns in imaging data
Accurate & reproducible assessments
Legend Risks Benefits Component of the topic Sub component of the topic