And Vijay Says…

Almost 7 decades have passed since Alan Turing asked “Can Machines Think?” . It’s hard to define what thinking means – and it continues to be hard today as it was in 1950. So Turing took a pragmatic approach and redefined the problem as “Can a computer behave like a human being?”.

It was absolutely a fine question to ask in 1950 because it was about the theoretical future and it provided a Northstar for AI researchers to guide their efforts . Now AI has reached a level where it is practical and the answer to the question has real consequences.

First – I don’t think machines think at all like we do . Most AI today is machine learning . What it does is do pattern detection at scale. And it takes a lot of information for the machine to “learn”. That in itself is an indication that machines are dumber than even little children today .

A toddler who has seen a picture or two of a dog on her story book can usually understand that the first real dog she sees is similar to a dog she saw in the book . And the odds of her mistaking a dog for a cat are minimal . That’s not true for machine learning at all – it needs a lot of labeled data to get to a comparable level if at all .

Second – we don’t just make decisions based on patterns alone . We use other information we know from the past and combine that with the newly detected pattern to make a decision . That’s not how AI works in mainstream today .

But Machine learning is amazingly powerful at a lot of things. It does not need to be like human thinking for it to be powerful and give us a lot of value . And like all of computing – it can also cause a great deal of harm. The time is ripe for us to get the debate on ethics of computing in general and AI in particular to actual actions with some urgency .

What is the purpose of AI ? Is it to replace humans at scale ? Or is it to augment humans ?

If it’s purpose is to augment humans – then isn’t it pure deception to make it pass Turing test ? Why should it solve problems in a way that makes an observer believe it’s a human ?

For example – if I ask a computer to add two five digit numbers , it can find me an answer in sub-second time . But to make me believe it’s a human that is answering – it needs to wait for few seconds or a minute to give me the answer . What is the real value delivered here ? And why should the virtual assistants have a human name and persona at all if the aim is to augment humans ?

Humans are imperfect and we are not always logical, nor consistent . That is our default mainstream behavior . Computers on the other hand have logical and consistent behavior as their default . It’s a waste of effort to make them behave like humans just to pass the Turing test . And obviously how ethical would such a system be that is fundamentally built to deceive ?

Computers can and should be used extensively to solve complex problems . They can do that without having to mimic humans just to trick us .

AI can do the world a lot of good – and that won’t happen if we don’t trust it . For example – there are well known ethical problems to solve for self driving cars . If we stop thinking about AI as human like – we probably will set a realistic bar for it to meet for us to be comfortable letting it drive our cars . AI enabled cars will also have many accidents – but we can have a logical discussion on “is it good enough that we can save a lot of lives lost through DIU and texting while driving by having more cars depend on AI ?”. If on a global aggregate – if AI can reduce the total loss of life every year , will we accept it ?

There will be many such difficult questions to answer now and in future . Can’t we give ourselves more room to solve them by taking away the need for AI to be human like ?

So should we free the world of AI to move past the Turing test so that they can focus on solving real problems without the burden of also deceiving us in the process ?

A client of mine gave me a very nice compliment today in a meeting – he said something like “You very naturally used the word colleagues instead of employees, and that is something very unique to our culture. How do you adapt like this?” . My honest answer was “After a long time spent in consulting, it is second nature for me to adapt to the client’s culture. I don’t do it consciously”. He suggested that I should blog on it, and I agreed – I think it might help someone if I shared my line of thinking here.

A good consultant is a change agent. I say that a bit “tongue in cheek” because for the longest time I have teased my colleagues who are experts in change management that “change agent is someone who can write an SOW selling change management” 🙂 . The primary value a consultant brings to a client is to help bring a point of view on how best to transform from where they are to where they need (not just want)  to go .

Change is darn hard. I don’t like change very much either. I like Carnatic music , and I like Indian food. It took a lot of effort for me to try Italian food and Western music.  I am glad I did because it exposed me to a lot of pleasurable experiences that I would otherwise miss on. But even today – if I have a choice, I revert to Carnatic music and Indian food 🙂

Knowing that I find change is hard for me – how do I influence change for someone else ? Let me go back to how I started liking Western music . It happened initially because a young aunt of mine that I was very fond of (I still am) started playing western music when I would vacation at her house. She and I had a lot of common interests and I got a feeling that if she likes it, maybe I should try it too. And I did. But the turning point for me was a lecture I attended in 2000 in Colorado where the speaker did a compare and contrast between eastern and western classical music. That helped bridge the gap for me significantly and I started appreciating western music a lot more.

That is roughly what I – often not even consciously – do when I am consulting. Every client has a unique culture. I try to understand and adapt that culture so that I have a good chance of walking in their shoes and seeing problems as they see them. It also helps build trust to a large degree. Once I have an appreciation of the problems and have built the trust – the next step is to step back and find a bridge from their current world to where they ought to be. When you build that bridge – or roadmap as we consultants say – then the future becomes less of an unknown. And when it less of an unknown, the resistance to change is lower and we can all move forward.

There is a small risk involved here that consultants should watch out for – and I have paid a price in my early career for not being aware of it. A consultant is only useful if they appreciate the culture of the client, but can remain objective about the end results one should drive to. It is quite easy – especially over long term relationships – to go native and think exactly the same as your client. The way I deal with it is to make sure that I always have a diverse team around me to keep me honest. Another strategy is to change roles from time to time so that you are forced to shift focus and learn something new .

One last point to finish my thoughts on this – often time is not your friend when it comes to getting to appreciate your client’s culture etc. This is one area where I think social media has helped me build empathy in quick time. I share my thoughts on a wide range of topics over linkedin, twitter and this blog. When I walk into a meeting with someone I have never met before in person – it is quite normal to expect that they have googled me and would have found some POV that I hold. Whether they agree with it or not – it provides an instant start to the relationship and I can build on from there.

 

 

It’s tax season now and I owe a bunch of documents to my CPA to file my taxes . It’s a pain in the neck to compile all of them . I do all my banking with one single company – and have multiple accounts with them . I can get a view of how much money is in all my accounts on the opening screen . But if I need tax documents – I need to go into each account and download them . There is no way to do a batch download or batch transfer to a cloud service .

It’s no different for mortgages and insurance policies and W2s and what not . I still have to go one at a time to each account to get the documents and manually compile it . Then my CPA has to look at each and put data into his software and file taxes. I am sure if this process were easier – I wouldn’t have to pay him a lot in fees , and he will make even more money than he makes today by having more clients.

I have gone through this problem when applying for mortgage as well – find documents one at a time, download it and then send it to the lender who looks at each PDF , and types the information to some screen and then someone takes a decision .

A (smaller) version of this problem happens when you join a new employer or start a side business.

There are 250 million or so tax filers in US . I would think there are at least 50 million mortgage applications each year as well . And each of them have to go through this time consuming and highly error prone process. Just thinking about it makes me want to scream !

What is funny (and sad) here is that vast majority of the companies – banks , mortgage lenders and the like – have some kind of customer 360 view for everyone they do business with . That is how they sell more and (hopefully) better service their customers . But these same companies, when they design their customer facing solutions, limit the customer to – at best – a dashboard that shows the combined view of accounts . It’s not done in a way that the customer can share information easily with the government , or a lender .

This situation is more than ripe for significant disruption . I would gladly pay for a service that can aggregate assorted documents for me as a secure cloud service – statements, tax forms etc – from the vendors I have accounts with . And then I should be able to just select the view I want to share securely with whoever I want – like the government , or a lender or my employer . These documents are all structured in a simple way – so it doesn’t even need to be shared in PDFs . An API based solution should be totally possible to provision .

Some combination of a company like Box and a company like DocuSign can solve most of this problem with existing technology and ecosystem . Tech is not the issue here at all – it can be solved by a range of tech from good old XML to the coolest blockchain and AI tech 🙂

Does anyone know if such a company exists today ? If not – I am seriously tempted to start one , or put some investment if someone wants to do it .