keesbuiel-blog

[keesbuiel-blog]

Difference Between AI & ML

Do you understand the difference between Artificial Intelligence and Machine learning? Interested to learn? Great, sit down or stand up and take a moment to read this short blog and we will provide our explanation of the difference between Artificial Intelligence and Machine learning. And, while we on the journey, we will take a moment to explain the TonkaBI approach to Ai?

Artificial Intelligence is two words which have two separate meanings, artificial meaning made by people, often as a copy of something natural or something that was created unintentionally. Intelligence meaning the ability to learn understands and makes judgments or have opinions that are based on reason. Sometimes, Artificial Intelligence in the computer programming world is misunderstood as a system or an application but this is not true, Ai is the study of how to train computers to perform tasks that share characteristics of human intelligence. Ai code can be embedded into applications, systems or devices this does not mean they’re Ai it means they’re augmented by AI.

Machine Learning, lets again look at the words separately. Machine - a piece of equipment to perform do a particular type of work. Learning - the activity of obtaining knowledge by studying it or by experience. ML in computer programming is when the machine can learn on its own without being explicitly programmed. Machine learning algorithms are used in sorting high volumes of data. Machine learning also supports Ai development by reducing the amount of hard rule based code, without ML you would have to write hundreds maybe thousands of lines of code within the Ai algorithm.

A computer (program) can gain the ability of Ai with little training and with a small amount of data, but this Ai will lack precision and intelligence in analyzing contrasting data e.g. computer vision. The Ai could identify a car door from selected images but not a car door from various angles or what side the car door is, and, it would not be able to identify a car door from random never seen before image data. Ai can further lack ‘intelligence’ through the wrong or weak algorithm and engineering ingenuity. A true Ai algorithm should perform its deigned task with high accuracy and scalability and outperform human counterpart at that given task, example AI image processing and classification in analyzing vehicle damage – faster, more accurate, multiple cases, and 24 hours per day – better than a human can manage.

TonkaBI uses two examples for the types of Ai we develop and use. These explanations and definitions may not be academically correct, but these are the terms we use in-house and when explaining Ai and the type of Ai we use when communicating to our partners and clients.

Narrow Ai – For TonkaBI this means the Ai is a decentralized standalone series of code that is embedded into client customer systems and processes. Narrow Ai does not have the ability to learn from data or its mistakes or its correct choices. It just performs a set task, day in day out to an acceptable standard. The Ai does not need internet access to work; neither does it require API’s. This Ai provides many businesses with the ability to have Ai technology in remote situation, embedded on hardware or environments where access to the internet is unachievable. The Ai also provides many businesses with a “Good Enough” approach to many situations. TonkaBI can update its Narrow Ai, which maybe embedded into client system, through “swapping” new code for the old. The new code would have been further trained with better ‘understanding’ ‘accuracy’ ‘features’ etc. This Ai grows in ‘steps’ as required for clients.

Active Ai – Similar to Narrow Ai but with a big difference, the Active Ai has a parent (a teaching model) that corrects mistakes and supports change and growth. This means the Active Ai can learn from data and gain knowledge that would be otherwise be forgotten or lost on the fly, in real time. So, the more data the Ai consumes the better it gets. Active Ai needs access through the internet or a network to the TonkaBI parent teaching module.

TonkaBI fully supports both Ai models and agrees there are advantages and disadvantages to both approaches. Our way of working and providing modern solutions to businesses has come from understanding our client’s what to have the code embedded in their own systems, ones they control and manage, our businesses model comes from listening to the market.

[keesbuiel-blog]

Blockchain or Hashgraph?

So, there we were, in a planning meeting, thinking about how to implement a parametric insurance product, how to fully secure and automate it....and then someone mentioned Hashgraph! What?????

Just as I was getting tired of hearing "blockchain this..." and "Lemonade that..." we now have Hashgraph....

What is Hashgraph? Why was it so important to mention and for insurance?

Distributed ledger space is a huge talking point for insurance – smart contracts, security, cost reduction, trust – it all points to a move on to this tech. But Blockchain come with limitations by design and are, by their nature, slow. By allowing the 'community' to come to agreement and throw away the blocks they don’t agree on speed suffers. In today’s world SPEED is king and this is a huge barrier.

Even in the slow paced world of Small Business Insurance, how many applications are there that can just do 5 transactions per second? None, I think. Hence blockchain have been all talk, to-date.

Many large companies across sectors are increasingly testing blockchain PoCs, but full adoption is constrained by core limitations such as transaction speed.

The Hashgraph algorithm, invented by Leemon Baird, the co-founder and CTO of Swirlds, is a consensus mechanism based on a virtual voting algorithm combined with the gossip protocol to achieve consensus quickly, fairly, efficiently, and securely.

YAY – QUICKLY – speed could be solved, and we can adopt.

“Hashgraph is an alternative to blockchain which is a first generation tech with severe constraints in terms of speed, fairness, cost, and security,” explained Mance Harmon, Co-founder of Swirlds and Hedera.”

For the layman what is Hashgraph? Well if you know what blockchain is (a distributed ledger technology) all about then this is relatively simple – Put simply it’s like Blockchain without the limitations.

Under test, Hashgraph has performed well. Since time is a trade-off between throughput, latency, number of computers, and geographic distribution, the tests demonstrate these trade-offs very well without adding in another trade-off. For example, the results show 30 computers can achieve 50k transactions a second across 8 global regions in 3 seconds, or merely 1.5 seconds across 2k miles, or .75 seconds in a single region. This is great news and it could be time to open up PoC’s to this new tech.

These speeds mean that processors such as Visa can think of building it into their current network and transaction speeds.

The Hedera Hashgraph platform is architected to address the market of distributed applications. The vision provides an initial three sets of services as the platform evolves:

1.       Crypto currency as a service for support for native micropayments.

2.       Micro-storage in the form of a distributed file service that apps can use.

3.       Contracts.

Great – how do we play and pay? It works through a platform coin — when you make an API call to one of these three services, a micro-payment is made to the company.

The Hedera platform’s existing technical framework is capable of anti-money laundering and know your customer compliance which sits well for insurance. Regulatory compliance combined with enhanced security and speed could signal a time to start.

With a secure, fast, public ledger, the future could include micropayments and massive-scale distributed p2p insurance and other new models could be developed. Match distributed ledger technology and AI and the future is bright!

So, what to do now? We are looking at this tech to architecture our insurance models. We have been waiting until now owing to the fundamental issues in existing blockchain processes.

[keesbuiel-blog]

Community Insurance

Millennials and Generation (GenZ) all have one thing in common - they don't like the idea of buying insurance. Wait...; maybe this is across all generations? Plus, there are the labels - lazy, entitled, self-centered, disloyal and spoiled, but again are this not all generations at this young age?

Let’s not focus on this, let’s take a closer look at this new (ish) group of young adults, what shines out for me is that they are turning to borrowing, sharing, renting and hiring rather than buying. They want the 'experience' not the hassle of ownership. They want the ease. Cake & Arrow have been studying this group and their insights are interesting, not just for insurance but for all of commerce.

Millennials have grown up with a fully functioning computer in their hand and this drives everything.

While not all Millennials/GenZ are alike, there is one term defining their generation most might agree upon – the convenience generation. From on-demand taxi services, to on-demand food delivery and on-demand entertainment, these adults have opted for new levels of convenience unlike any other generation before them. Millennials and GenZ are choosing the convenience of Boxed and Amazon Prime over the prices of Costco and Sam’s Club.

How is this insure industry responding? We see terms like “digital transformation,” “personalization,” and “on-demand,” these are the chosen jargon of the industry, used in industry events, panels, webinars, and reports across the world in an attempt to set clear objectives and guidelines for companies that want to stay relevant. What’s more, these terms have been portrayed by others as promising opportunities for those willing to embrace the journey.

But is this true? Do we really know that this is what drives a millennial insurance shopper? Or is that just the convenience of the ability to deliver on this in the context of what a company was doing anyway? Let’s make it digital??? This is the easy option; it’s got to mean something. Having a fancy front end, is this enough?

I believe that digital is a step in the journey but it not the end of the journey or even the point of the journey. The journey to me is much more than this, it is about making a difference and bringing out the true qualities of the industry, the fundamental reason for insurance and the difference it can make to those who take it out. These qualities are hidden and pushed way back in the dim past of the industry and the companies that form the industry. It’s the WHY of the industry, yet this seems lost.

For Microinsurance, we are going to bring out this WHY. People talk about Peer to Peer insurance, but isn't this what insurance is anyway? Why do people want insurance, why is it such a good thing? Is the only reason you buy insurance is because you are forced to? Let’s look at a bit of the origins of insurance to find out the why it came around in the first place....

 The first methods of transferring or distributing risk in a monetary economy were practiced by Chinese and Babylonian traders in the 3rd and 2nd millennia BC. Chinese merchants traveling treacherous river rapids would redistribute their wares across many vessels to limit the loss due to any single vessel's capsizing. The Babylonians developed a system and practiced by early Mediterranean sailing merchants. If a merchant received a loan to fund his shipment, he would pay the lender an additional sum in exchange for the lender's guarantee to cancel the loan should the shipment be stolen or lost at sea.

It's about looking at activities, groups and risks and providing cover for them should something go wrong. The whole point of insurance is to build a community and allow large groups of the members to pay a little (insurance policy) towards the losses (claims) of the few. Even there is more innovative concept like Climate Change Insurance in this challenging atmosphere when we are facing such a anthropocentric situation. Even Parametric insurance is a type of insurance that does not indemnify the pure loss, but ex ante agrees to make a payment upon the occurrence of a triggering event.

This is the thing that seems lost today, users don’t’ see their part in the community giving benifits to the few. Today’s insurance industry has drained all the feeling and warmth from their products. They forget the people paying for the few who claim. They forget to highlight how the easy payments of claims really benefit the ones who have suffered.

Microinsurance is all about the claim. Building in transparency to all customers. Not just the claims but the policies - allowing the communities that are built around our policies to see the claims and see how their money has helped support the community.

GenZ and Millennials want to be part of communities and helping support others and bring sence in to thier lives. To grow and be relevant to these generations’ insurance needs to update it and be more than digital, it needs to get back to its early existence and be part of the community, for the community.

Know more about us at: http://blog.microinsurance.com/post/climate-change-insurance

[keesbuiel-blog]

It is widely known that the United Kingdom was involved in the early research and development of the internet, although the United States of America was driving the project and provided the funding; the UK played a crucial part. Tim Berners-Lee, Peter Thomas Kirstein and Donald Watts Davies are some of the names that stick out when you talk about the development of the internet. I wonder what they would think about how we use the internet today.

To help answer this and other questions such as what country produces the most traffic? How fast is our internet speed? Are we at the front of the internet advances or the back? Let’s having a quick look at what the average person uses their phone and the internet for.

The top 5 activities for mobile devices are: 1 map, 2 instant messaging, 3 music, 4 photos and 5 weather. You could say we all like to find a nice place to take a good photo while listening to music and then tell everybody about your activity and today’s weather.

Despite the UK having one of the fastest and longest physical internet lines in the world (England to USA) I find it hard to fathom why our internet speeds are so slow and why our ISP bang on about speed so much when in reality the UK is badly served. Have you ever connected to the internet in the UK? It’s really slow. If your living in the UK I bet you have experienced video buffering when streaming and I bet if you have ever complained to you IPS provider they tell you to turn your router off and on again or the buffering is due to peak hours. I think this is a load of twaddle, peak hours? We are in 2019 not 1999.

As of January 2019 the top 5 countries with the fastest average fixed line internet speeds are: Singapore with 189.38 MBPS, Iceland with 147.13 MBPS, Hong Kong with 139.58 MBPS, Romania with 107.42 and South Korea with 103.51, these are all great speeds to be averaging in 2019. The United Kingdom on the other hand averages a miserable 55.14 MBPS this is not a good speed to be averaging in 2019. I didn’t expect Romania or Iceland to have such a good fixed line average. The top average mobile internet speed range from 63.13 MBPS to 48.64 MBPS and again the UK does not appear on the top 10 list. Forget BREXIT, this is INTERNEXIT!

In Singapore you could download and fill up a 120GB hard drive in around 10 minutes that’s around 31,000 songs. In the UK it would take around 36 minutes with a download speed 55.14 MBPS but if you have a connection speed like mine it would take over an hour! Forget video and pictures!! Slow internet speed is a concern for e-commerce companies and anyone offering a web based service. 40% of all people will abandon a website that takes longer than 3 seconds to load and 80% of all those people will never return. This won’t be solely down to website optimization as internet speed will play apart; if the site you’re visiting doesn’t load you typically abandon the request and try again or try another website to see if that site loads. We briefly discussed this in a previous blog. This means that you cannot adopt the latest video formats and designs or up the graphics on your site – you are constrained by the ISP!! Basically you have to down size everything on your home page or landing to ensure the page will load within 3 seconds.

In the UK online shopping is big business the online shopping market per capita is around $4,000 which is pretty astonishing when you compared the UK internet speed and usage to USA. There must be some patient shoppers in the UK. I sourced a list of the apparent top websites visited in the UK. Considering online retail shopping brings in some much online revenue it doesn’t seem to bring much traffic. The top 20 Google- YouTube - Facebook - Amazon - BBC Online - Wikipedia - eBay UK - Reddit - Twitter - Live - Netflix - Twitch TV - Instagram - Office - Yahoo - Pornhub - The Lad Bible - VK - The Guardian - Live Jasmin. If you look closely you will see a lot of free service providers who use their free website space on their platform for marketing. This product marketing is done through your search history and cookies.

TonkaBI builds capability to process insurance claims, such as car insurance claims, Ai Image Processing though Artificial (AI) and Machine Learning. TonkaBI’s Artificial Intelligence capability is based on image processing. This is used to build a capability that is specific to your company and process.

It’s reported most people in the UK prefer to use a desktop rather a mobile device or tablet, care to take a guess at the sites most visited on a mobile device or tablet? Around 51% of people use desktop computer or laptop when surfing the web this fallen by 3% in recent years whereas 36% is from smart phone this is expected to grow. I know you don’t need fast internet to purchase clothes etc online but considering the market value VS internet speed this is quite low. However new VR tech is on the way and augmented sales channels – how will the UK stack up then?

By 2021 it’s predicted there will be 3.8 billion active Smartphone devices and with Asian owning half of those devices. 61.09% of all web traffic from Asian is from mobile devices whereas Europe’s mobile web traffic only consists of 37.08% to the overall internet usage.

Over 20 years ago, in 1992, global Internet networks carried approximately 100 GB of traffic per day. Ten years later, in 2002, global Internet traffic amounted to 100 Gigabytes per second (GBps). In 2016, global Internet traffic reached more than 20,000 GBps. That’s an astonishing leap. Check out our earlier blog on data.

Know more about it at: https://www.tonkabi.com/artificial-intelligence

[keesbuiel-blog]

Insurance innovation

Why should the insurance industry care about innovation? Their products are about claims, how can you stop those with innovation? They are part of the product, right? Well no, typically the insurance industry, as all businesses, cares about all its costs, the complete business model. The insurance industry has a very public metric, the 'combined ratio' or, I would say, the generally poor combined ratio. This metric should be the driver for all innovation

So what is the combined ratio? This is typically expressed as a percentage. A ratio below 100 percent indicates that the company is making an underwriting profit, while a ratio above 100 percent means that it is paying out more money in claims that it is receiving from premiums. The combined ratio is the adding up of all the costs of the industry.

This is the driver behind the insurance industry - it’s like the lap time of a F1 car - shed a 10th off here or there and you are in the lead and seen as the winner. Don’t work on the lap times and you are seen as going backwards.

The industry 2017 combined ratio was around 105% and in 2016 it was around 101%. Meaning that overall the insurance industry lost money from its core business in both years and 2018 will be similar.

So you would think that this would be a huge RED ALERT. No, this is business a normal.

The industry lives in a world of underwriter profits. This could be seen as very strange to other businesses. The industry struggles to focus on their main core driver because profits are driven from investing the insurance premiums - not driving the costs out. Imagine if F1 teams did not care about lap times? What would that race look like?

However this is the opportunity. Focus on the combined ratio, use Supply Chain Insurance to drive costs down - the ability to affect this ratio in some way through speedier sales, smarter claims, better intelligence in underwriting, cost cutting through the business process, straight through processing, etc. They are huge wins to find.

The current business model is stuffed full of issues and problems that are all fully baked in to many of the incumbent companies making change hard. Issues and Processes such as: any claim starts with a fraud review; the broker or agent selling the policy does not deal with the claim or carry any of the risk; the agent, broker, MGA, Underwriter, reinsurance business model is too expensive - each taking %'s to cover their costs; within in agents and broker businesses the pay policy fights against innovation; data is not collected in one place, its dispersed across the supply chain; the Policy is mainly a list of exclusions and complicated, it wants a court fight; it’s hard to get new risks covered or understood; micro policies are too expensive to process; and this is to name only a few of the core issues within the existing model.

You get the picture, insurance is not efficient.

The world of Small Holding Insurance is trying to come up with new ideas and models to improve upon existing structures. However, these tend to be around the edges of the problem. What is needed is a drains-up overhaul of the process and a move towards true digital insurance companies - ones that own the customer and all process straight through, from marketing to sales to claims and on to renewal. There are a few new companies emerging, ones that own the whole process, Ping and ACKO.

The collection of data and the ability to speed up and take the friction out of the process is essential if insurance companies are actually going to make a constant profit from insurance and not rely on investment income. The aim needs to be a constant combined ratio of fewer than 100% whilst keeping the policies priced sensibly. It is possible.

Data and the ability to process and learn from the data are essential. Insurance companies say they have loads of data. This is only partly true - they generally have very little transactional or process data covering the buying cycle of the policy and limited marketing data. They have very few insights into their customers, their motivations, etc. This is because many policies are sold through partner channels - for example if an agent is selling the policy - they don’t pass on or even collect data on customers who looked and did not buy - they certainly don’t find out why they looked and did not buy. This data is lost to the insurance buying cycle.

These simple examples explain why innovation is so hard. The larger companies are changing - they are going more direct, they are doing deals that embed insurance into products, they are starting to build their brands to be direct to the customer.

[keesbuiel-blog]

Voice First Tech

When you need to solve a problem that involves another person, what’s the best way to solve it? By having a conversation. We’re finally able to communicate with our devices—phones, computers, wearable devices, smart speakers and more—the same way we talk to one another when we need to get things done.

We are approaching a time where we move from typing to speaking to our devices - Voice First. Voice first is here now and being integrated in to all the speaker and voice devices and processes.

As this tech becomes more accepted and user friendly, hiding much of the complexity of voice-enabled technologies, users will use their voice to order groceries, book a cleaner, call a cab (or give commands to a self-driving one), make appointments, remotely control their homes and more and more - I mean we use our voice for loads of things already - it will just be integrated into our tech.

Will this be the 'final' single interface for interacting with a diverse variety of devices at home or on-the-go, making it easier for users to accomplish any task hands-free? I don’t 'think' so, we will have to 'think about that' first.

However voice-first will be a huge step forward and change in the way we connect and control and ask our devices to help us.

Already the comedy has come - check out this - this is just an example of how results will be given for different parts of the world or parts of the country. It’s funny but also has some truth in it!

There are a variety of examples of voice-first tech and use cases; this is a great demonstration of the possibilities today showcased by Alexi team

There is a rising importance of Voice Analytics and Voice-First. Many of us have seen the adverts for switching on security or locking your doors through voice and we maybe seen the movie Her.

However a few of us in the globe are just becoming accustomed to the technology. And as usual many users don’t really understand the tech and can’t use 80% of the features - the point is that even just a few features are convincing user to discover this tech and start using it.

As more people use and develop tools and systems, analytics will help developers better manage expectations, anticipate errors and help users discover features relevant to their needs and drive up adoption of voice-first like Ai Image Processing.

In developing voice tech and text messaging tech there are numerous metrics that can be tracked by developers for voice platform improvement and development. These include:

• Sentiment analysis: The language used, as well as tone or pitch of users, can help to measure the sentiment of a brand or of the user experience itself helping direct support and improve engagement.

• Intent and parameter: Intent (what the user is asking for, like “when is the booking?”) and parameter (the specific, contextual request, like “how late are they going to be?”) provides insights on consumer behavior.

• Pathing: Pathing includes the steps of the conversation as well as what actions users take afterward.

• Errors: Errors and null statements make apparent where devices and platforms fail their users.

Voice and text applications will become a key function as a part of brand omnichannel strategies - the ability to move with the customer as the customer changes platforms or locations etc.

TonkaBI is the expert in Innovative Data Technology, especially for the Insurance AI Industry! We understand what challenges insurance companies like yours face on a daily basis and we’ll be with you every step of the way, working with you to make this process as effective and as efficient as possible. TonkaBI is a living breathing organism, in all aspects, and if something needs to be adjusted, reorganized or rebuilt, this is something TonkaBI knows how to do this, do it right and do it quickly.

[keesbuiel-blog]

Invest in technology and bring value back to your business

TonkaBI is supporting insurance companies today in the race towards a smarter future for general insurance and claims processing. We all know that artificial intelligence is here and it is here to stay. But do we have a clear idea as to how fast it’s coming and its real true capabilities? TonkaBI has worked in AI and IoT and Machine Learning, mainly focused around insurance analytics and is able to share its experiences and techniques with interested insurance companies, companies that want to take advantage of this new technology.

Machine learning automation delivers a high level of efficiency that isn’t humanly possible and will increase query response time by a 100%. The most mundane tasks within processing and within business systems can be carried out by embedded AI (Artificial Intelligence). This has been proven to be highly effective and more efficient than any human. Machines work without breaks - 7 days a week, 24 hours a day and no holidays. That’s being productive!

What your customers experience from their perspective can be illustrated through distress signals. These distress signals can be taught to machines through machine learning algorithms. The machine will then identify these small issues in real time, allowing companies to take immediate action for the resolution. Watch, report, respond and manage all in real time giving you the ability to resolve small customer service issues before they build up into a big issue.

How is this done?

TonkaBI will complete a structured audit and scope of your requirements and how your company operates and processes data to gain insight into your business. This will be crucial for when the project commences. More importantly this will help TonkaBI understand you and your company and during this time what you desire from the technology and find important, will be highlighted and expectations can be laid out. Communication is the key to success through-out the whole project.

Once the project scope is set and the audit is complete and all parties have agreed upon the delivery TonkaBI will start a PoC (proof of concept). The aim of the PoC is to again understand and gain knowledge, furthermore, to start implementation of the ML algorithm into a small sector of your business, to provide proof that machine is indeed working, adding value to your business and also expectations are getting fulfilled.

TonkaBI’s team will be continuously working with you every step of the way to improve and optimize the code. This is all a part of our service and determination to bring you the best results and value into your business.

Once the machine has spent time in your system and the algorithms have adapted and gained insight this is where your will start to see a ROT (return of investment). As mentioned the biggest plus factor for machine automation is the ability of continuation at rapid rates, no breaks, no holidays just 24/7 work!

Every TonkaBI MI and AI solution is unique and one of a kind, we create solutions for software.

[keesbuiel-blog]

Hacking small and medium businesses - why bother?

Why would a cyber attack on 15-person Company raise any alarm bells? Why is it a concern to anyone let alone the Department of Homeland Security?

There are millions of small businesses in the USA and globally, in fact over 100 million of them and this number are growing each year, and faster as the world changes employment strategies.

So, why are any of them a concern to a government agency like the DHS? Surely the DHS is focused on banks and larger companies, this is where the threat is? Keep them safe.

No, this is not so true. In a recent example, a small business working with utilities and government agencies, suffered a cyber attack that was an early thrust in the worst known hack by a foreign government into the USA's national electric grid.

The Wall Street Journal has reconstructed the events around the hack that revealed huge vulnerabilities at the heart of the electric power system.

Rather than strike the utilities head on, the hackers went after the system’s soft and unprotected contractors and subcontractors. There are hundreds of them, all vulnerable and some more than others.

This should be sounding an alarm bell to every large corporate if not every company.

Small and medium business generally has no reason to be on high alert against foreign agents 24 hours a day. Why would they be? They also don't have the people, systems or solutions in place to do this.

Yet through these small companies the hackers, in this case, found the footholds necessary to work their way up the supply chain. Enabling the final target to be reached, hacked and exposed. Some experts believe 20 or more utilities ultimately were breached.

The hackers have the time and resources to do this and they are aware that small and medium sized businesses are a very soft target.

The WSJ article is a must read, I am not going summarize it here to save you time- just read it!

Have you read it yet?

On a similar note and to underline the issue the FBI is investigating the alleged theft of 18,000 insurance and legal documents relating to the September 11 attacks on the World Trade Center by a hacker with a long record of holding companies to ransom. This ransom attack, if it did happen, highlights the vulnerability of a business not just from within but across a huge web of suppliers and partners.

This type of breach, Data Breach can lay your clients details bare, data lost and cause untold issues, at the very least a PR nightmare.

Where does this leave you?

What can you do about the growing threat of hackers? First, put in place the best tech barriers you can afford, get some advice to - know where you are weak. Buy cyber liability insurance to cover the recovery costs too. Vulnerabilities change all the time, insurance is there to bring you back to life when all else fails.

Then patch your biggest vulnerability: your people. They need training and awareness of these issues, especially if you work for large corporate or government bodies.

It’s not just about employees having smarter passwords and spotting sketchy emails but also to think about their online actions. This is not about a list of rules; it’s about awareness and responsibility. Remember rules create a path for hackers to follow....

[keesbuiel-blog]

Micro Insurance by STP

The generally accepted definition of Micro Insurance is the protection of low-income people (those living on between approximately $1 and $4 per day) against specific perils in exchange for regular premium payment proportionate to the likelihood and cost of the risks involved.

This definition seems to focus on one Target Market - the low income peoples of the world. This target market does not typically buy insurance and are generally ignored by main-stream insurance companies.

Here at Microinsurance we do not really like this definition, why pick on a population with a particular insurance sector?

We prefer a wider more inclusive definition of Micro Insurance, one that allows products to be developed, sold globally, works in the new Sharing Economy Insurance and includes solutions for many of the issues surrounding selling, distributing and managing micro insurance policies and schemes. Products that cover all populations based on their specific needs.

The problem with the accepted definition for Micro Insurance is that it is exactly the same as one might apply for regular insurance except it’s aimed at low income people. I.e. Insurance is the protection of people/businesses against specific perils in exchange for regular premium payment proportionate to the likelihood and cost of the risks involved.

I guess the general insurance industry could not bring itself to call it Low Income Insurance, like they do for High Net worth Insurance so the term MicroInsurance was adopted.

Anyway we are getting away from the point.

Micro Insurance has many challenges and these challenges are not just issues surrounding products for low income families. Just like the definition for Micro Insurance is the same for any insurance, so are the challenges for any insurance product when faced with high volume sales and policies, new Business Insurance models like the sharing economy and platform businesses.

Today’s insurance industry is not very well geared up to deal with high volume sales and claims. The nearest you get to high volume is car insurance and in terms of micro insurance, these volumes are very small.

Here at MicroInsurance we see that the investment of resources needed to solve the challenges of micro insurance can be used globally for insurance policies that are simply small, micro. That can be used by everyone. Policies that match user and policy.

Typically insurance policies are complex and expensive. Insurance companies must like these as they sell millions of $$ of them each year.

We look at insurance the other way. We think of making insurance simple, event driven and the policy value small. Covering events that might last for a journey, a purchase, a short period, a job etc, hopefully you get the point – rather than buying for a year or month etc, cover the event instead.

Simple. Tech Based. High Volume. Embedded. Transactional.

Simple

Insurance policies are generally complicated - many insurance TV adverts point this out, focusing on saving money, making the process simple but hiding the complexity since people tend not to read 'small print'. The industry has this issue in its DNA. They are contractual documents after all.

Micro insurance policies do not need to be complicated, times needs to be invested to simplify the whole process ensuring that policies are fully incorporated through the process - marketing, quoting, buying, renewal and on through to a claim.

Tech Based Leverage in the tech in your phone or on your PC to good effect. Linking the process such that the customer journey is well thought through and connected, end to end, right though the customer journey and the life of the policy. Processes are built with APIs and integration at the core of the tech to allow.

High Volume

Insurance companies generally do not like high volumes of anything - especially claims. They simply are not geared up to deal with high volumes of customer contact for sales, queries, claims and complaints. They typically pass these tasks to others - Sales via aggregators or agents and brokers - Claims are passed to Third Party Administrators, specialist claims companies - Complaints are pushed overseas to keep costs down. Insurers and brokers split the process across many companies and struggle to have a complete view of the customer apart from financial performance and product based metrics.

To manage high volume required by Micro Insurance means owning and investing in the process and managing the transaction end to end. Entering data once and then straight through processing along the entire journey. This has the advantage that lots of data is collected and allows for better data usage and management which leads to improved process, more customer engagement and pricing.

Embedded

Rather than buying policies for Cars, Gadgets, Home etc more and more insurance will be embedded in the process and by your use the benefits of the insurance will be passed on to you. Home security and Home help devices could come with home insurance, electric cycles would be insured against damage and theft, App that allow you to use the cycles could have insurance added per KM and variable depending on if you are in the local park or on a busy road. IoT devices for crops could come with insurance that monitors the crop and the rate varied depending upon the actions of the farmer and the weather.

Transactional Insurance does not have to be on an annual basis. The current process is to some extent driven by the inability for insurers to manage volume and customer engagement, it’s cheaper and easier to manage once per year rather than 12 times a year or on each usage – say 1,000 times a year. Imagine if car insurance was all usage based? This would be fair; the way insurance is managed would be very different. This is the world of transactional insurance – insurance when you want it and no more. High volume, small value insurance policies based on the transaction. Managing the policy life cycle, monitoring and claims fully automatically and on a transactional basis. Micro Insurance based on activities and usage. The distribution model and commissions for brokers, agents and partners all built into the process and a transparent claims process that has clear triggers for payment. Examples of this are parametric insurance for travel, hurricane and agriculture.

[keesbuiel-blog]

Insurance, Blockchain & Oracles

Here at Microinsurance we continue to explore blockchain, the world of “a smart contracts”. It is our current belief that ‘smart contracts’ will be the core insurance uses-case for blockchain. Why is this? Because the MAIN thing that people buy when they take out insurance is a contract. Nothing more – insurance is a contract to pay if a set of events occur in the future and is within the bounds of said contract.

Today, there is nothing wrong with the contracts that insurance companies use, well they are wordy and fully of small print, allow for ambiguity and generally need lawyers and administrators to navigate the many escape routes that are built in. So to introduce 'smart contracts' into the process takes tech and investment - two things that the insurance industry is slow to do.

Insurance contracts should, on the face of it, be very simple. They should ensure that each party is dealt with fairly and each party receives and gives what they thought when it was signed. There are many instances where insurance companies fight claims and slow down payments. This is where smart contracts can fulfill a need. Welcome to the world of smart contracts, these contracts don’t leave the chance for interpretations. For insurance they enable so many benefits.

The use of these contracts are now linked to the rise of blockchain and a change in technology and process that is still very young for many companies and, in many cases, at a pre-Proof of Concept (POC) stage in insurance companies.

The use of blockchain in insurance is the tech game changer that is needed within the industry to bring straight through processing advantages to insurers and brokers. Smart contracts will, eventually, be used for sharing economy, gig economy, IoT and platform insurance. Making the process very transactional.

The main idea behind the use of ‘smart’ in smart contracts is to use tech coding to determine the relations and obligations between parties and automatically administer these clauses and relationships. The contracts make possible to exchange money, property, shares, or basically anything of value in a transparent and non-conflicting way.

Basically, the smart contracts have the trust built in. Add this to the idea of a decentralized blockchain network and you can start to see the power of these contacts within the insurance industry, especially around parametric and transactional insurance. The conflict of ambiguity is removed. The power of speed and volume enhanced.

The term “smart contract” is widely associated with Ethereum or IBM. Currently the Ethereum smart contracts are the most popular. However, it is possible to create smart contracts on other Blockchain platforms.

In 2018, a US Senate report said: "While smart contracts might sound new, the concept is rooted in basic contract law. Usually, the judicial system adjudicates contractual disputes and enforces terms, but it is also common to have another arbitration method, especially for international transactions. With smart contracts, a program enforces the contract built into the code."

Other forms of smart contracts are Ricardian contracts. This form of contract maybe more relevant to the insurance industry. A definition, from its creator, says a Ricardian contract is “a digital contract that defines the terms and conditions of an interaction, between two or more peers, that is cryptographically signed and verified. Importantly it is both human and machine readable and digitally signed”. This definition makes it very usable for insurance where both parties may want to use the contract from time to time.

A Ricardian contract registers a legally valid and digitally connected document to a certain object or value. A Ricardian contract places all information from the legal document in a format that can be executed by software. In this way it is both a legal agreement between parties and a protocol that integrates an agreement offering a high level of security because of cryptographic identification.

With a smart contract, a person could, for instance, have a hurricane insurance policy contract that is encoded in the block chain in the form of a set of rules.

In case of the hurricane coming, the smart contract could then automatically transfer the claim money to the beneficiary. The Insurer may provide additional constraints, such as enabling the transfer only when the hurricane reaches certain intensity and tracks to a location within certain parameters etc.

Since smart contracts’ conditions are based on data stored in the blockchain, they need only to rely on external services, which take data from the “real” world (e.g., from hurricane tracking and location tech) and push them to the blockchain (or vice versa). These services are referred to as “oracles”. By considering this example, an oracle could inspect the presence of a hurricane’s track and intensity to identify whether the person/company (Insured) is eligible for a pay-out. This eligibility could also test against claims materials instantly read on the blockchain such as invoices or other records. This would then trigger an instant, automatic, payment.

[keesbuiel-blog]

Today's Business Tech

You’re sitting in your office chatting away like there’s no tomorrow with your AI powered computer; you are analyzing risk and predicting the future of your company to some extent. Your AI powered computer is projecting 3D interactive data models onto your desk, you are pulling apart the data adding new dimensions and various layers, it’s all going well and the future looks bright. While doing this you ask the computer questions like “show me the value of the risk of insuring commercial property against flood damage within America and the state of Florida and within zip code 33187 and how I can increase my profit by 10% on the value of property between four hundred thousand and nine hundred thousand dollars”. The Ai would then take those values and run equations to produce the answer, boom! What would have taken you hours is accomplished in minutes. Soon you will be able to analyze any risk or anything about your company by simply asking the right question, easy work. All the boring stuff like writing documents, looking through data, building charts, etc will be done for you, basically all the stuff you really don’t want to do but you have to do, the time-consuming stuff.

Most of us have probably seen videos or read articles on the internet about how AI is evolving and how AI is going to replace many of us all by AI powered robots or even drive us to work or a drone is going to deliver pizza to your front door. These videos and articles are very fascinating, even entertaining, to read and watch. But we are a long way off integrating them into our business and everyday lives.

Here, I want to separate the dreams and hype to the reality and what is available for your business today. Many people have seen the technology they want, seen the video, gone to the presentation. They know what they need for their company but don’t really understand how to start on the journey. I want to help you start the journey.

Picture this, at the moment your contact center team is 20 strong with an average waiting time of 8 minutes depending of the time and day. Your business is growing and so is the demand from your customers for better service. The traditional approach to tackling large volumes of call is to hire more people, outsource the calls or restricts the time customers can call; none of these options are scalable or good for business. To add to the problem customer service agents have to capture the details of the call after every call which adds to whole problem. Customers call up for many reasons such as general queries about your product or service and information that can probably be found on your website. These types of calls last only a matter of minutes, but they add up to around 45% of all call volume, what a pain. Plus, you have a 15% churn on your staff and training takes between 2 and 3 weeks.

How great would it be if you could lower the cost of your call center, increase customer service and support your customer 24/7? Well this is completely possible and many companies around the world have already put this in place or are tackling this issue by integrating a chatbot into their business. Chatbot can

Reduce     contact center pressure.

Customer     support targets are met.

Increase     customer service hours.

Reduce     contact center cost and over heads.

Higher     customer satisfaction rate.

Lower     waiting times.

The list goes on.

A chatbot maybe not be able to understand all your customer problems and answer all their question today but the chatbot will be able to in time and this starts by integrating a chatbot into your company to handle a % of customer queries. A new chatbot process will allow the customer service team to drastically improve their efficiency and customer satisfaction. The team will be more motivated, and productivity will be at an all-time high.

Have you ever made an insurance claim? Do you work in an insurance company or own one? Maybe none of the above but what you may or may not know is the insurance claims process often involves analyzing huge amounts of images. These images can come in different formats like PDF’s, photos, videos and scanned documents there might be more but these are the most common. This isn’t just about insurance; any business that processes images can use this tech.

If you or anyone processes hundreds of images a day or week be honest, is this a good use of company time and resources? Is analyzing images manually cost effective? Your answer should be no. The biggest challenge people face in analyzing images as a part of their job is to stay motivated and to maintain a high level of concentration throughout the day while dealing with the same repetitive tasks. Looking for fine detail within images is tough. How do you take on more work and speed up processing time? Is the only way to hire more people or to outsource? No, neither solve the problem just add to it and increase the risk of error. Today there is an answer.

Artificial Intelligence, AI can be used to analyze images of all shapes and sizes and against all types of detail. Once the AI application is set up you will be able to earn a fantastic return of investment through many channels of your business. The beautiful thing about technology is it doesn’t sleep, take days off, get married or get sick, technology is a work horse which works 24/7 all year round and is designed to improve quickly. Analyzing an image or a set of images with AI is fast, very fast; the AI application can pick out key information and extract the information in a matter of seconds even the finest detail will be picked up. A good example of how fast and accurate AI Image Processing is, take 50 car images and time yourself to see how long it takes to list out the make and model of every car you identify. Do this 10 times a day and see how board you get and then ask yourself was this time well spent? Technology like AI is meant to do these types of repetitive and time consuming tasks.

Benefits to Your Organization

Drastically     increases the number of images and videos that can be processed per hour.

Processes     images with a greater degree of accuracy.

Reduces     labor costs.

Removes     human error.

Having technology such as AI image processing replaces humans is this type of process isn’t a negative it’s a positive. Companies and people can reinvest their time, money and energy back into back into themselves or their business for a brighter and more positive future.

[keesbuiel-blog]

Artificial Intelligence Image Processing

Artificial intelligence image processing combined with machine learning is making huge leaps and bounds within the insurance industry. The machine learning algorithms will one day replace all human processing within the insurance claim center industry; it’s already ready happening with chat bots and soon machine learning AI photo recognition. I think we all can appreciate when artificial intelligence combined with a well-developed machine learning algorithm out performs any human with repetitive work such as processing photos of damaged cars and insurance claim processing. TonkaBI hopes to break into the car damage claim processing market in India and Africa by machine claim auditing and AI photo recognition and image processing. TonkaBI has already started working on the technology which we hope will revolutionize the way insurance claims are processed. TonkaBI will bring the current claim processing time down from 2 to 6 + hours to under 5 minutes.

The challenge

The car insurance industry are facing a prolonged time in estimating quotes and finalizing claims through their claim centre the process which ends up back logging all other claims which increases work load on the insurance provider and frustrates their customer, it’s like a never ending cycle.

The current process outline:-

A Policy holder uploads images of their     damaged vehicle to insurance providers mobile application or web based     platform.

Insurance providers claim centre employees     evaluate the images and contact the policy holder for further clarification     of the damaged vehicle and the photos.

Insurance provider’s employees then calculate     and give the policy holder an estimated repair quotation based off the     photos and from previous claims from the same vehicle.

The Goal is to take the current process with all the time and effort and reduce this down to a few minutes and considerably less human intervention to provide the policy holder with an accurate quote for the repair of their vehicle. Insurance providers will also be able to clear all back log of claims. Insurance companies will be able to increase revenue and customer base, allowing companies grow fast and economically and to deliver very high levels of customer service.

TonkaBI fully understands the Use Case and can deliver an automated process that is managed and controlled through machine learning and artificial intelligence technology.

Summary:

Artificial intelligence image processing combined with machine learning is making huge leaps and bounds within the insurance industry.