The risk and reward of using AI in stock market

Got invited to a talk on “A.I. for wealth creation in the stock market” at an investor’s meet in my hometown and came away completely bamboozled. The speaker, the CEO of a brokerage firm, seemingly an expert on stock markets had zero knowledge about how AI works. He threw up a few slides on AI that were incomprehensible to the largely local non-tech savvy attendees and then started demonstrating ChatGPT as the way of using AI in the stock market to make better returns. But when he was asked to query ChatGPT to predict the next day’s stock market outcomes it couldn’t.

AI is mostly being used in automation and prediction now. Fundamental to AI is data and data analytics which leads to predictive analytics. The core aspect of AI is machine learning (ML) and there are two basic types of learning – supervised and unsupervised learning. All predictions from continuously changing data such as forecasting stock or trading outcomes and market fluctuations is through supervised learning. Pattern recognition is how AI systems identify patterns in data and then use those patterns to make decisions or predictions. One of the prevalent algorithms used for more accurate predictions is the backpropagation algorithm.

Now, an ideal investment portfolio would be a combination of high risk high reward investments (stocks, mutual funds) and low risk low reward investments (government bonds), the low risk ones to offset potential losses from high risk ones. Better AI prediction models can increase investment in high risk stocks leading to higher gains. But human sentiments and emotions play a major role in stock values and we are prone to mass hysteria. AI has no emotional intelligence so AI prediction models cannot factor in data on human emotions and can consequently lead to huge stock market losses.

Why we need to understand our origins and history before building artificial intelligence

According to ancient Sumerian tablets, a race of beings from another planet called Anunnaki descended on Earth in search of gold. Why they needed gold isn’t clear yet but they needed lots of gold. When the slaves they brought with them to mine for gold revolted, they extracted the genes of pre-human race of beings (Neanderthals, Homo Erectus) and combined with their own genes to create us. This is up for a lot of debate and in Genesis 1:26 in the Bible, God says “let us create them in our own image” where “let us” means there is no one God that created us. I am a believer in this story for two reasons:

1) Our natural tendency to look up when we pray which indicates the Gods we are worshiping came from above us

2) Unlike metals like iron, copper and aluminum, gold is useless to us. Then why take all the effort to mine it and why is it precious to us? Tons of gold have kept in a temple in India under the protection of the temple God for hundreds of years. This means gold is important to the Gods which is why it is precious to us.

Most importantly, the story is an indicator to us that we were created as biological robots to obey our creators. Why didn’t the Gods simply build mechanical robots for mining and why did they take all the effort to engage in genetic engineering to create us? Because they wanted us to think on our own and also have emotional intelligence. Why emotional intelligence? Because they wanted to control us emotionally and not mechanically. How? When we talk about Gods, we do not ask ourselves how did the concept of God and the fear of an unknown all knowing entity became embedded in us. We are in the 21st century and with so much of technological advances we are still under the control of Gods through various religions. There is nothing more advanced, complicated and powerful that can control us than the concept of God.

Do we have a similar controlling mechanism for the intelligent machines we are building? What would happen if AI grew to a point where it could think and act autonomously? The scene in The Matrix where Agent Smith tells Morpheus that human beings are the most dangerous virus on the planet has stuck with me. It is not only true that we do behave like viruses, if intelligence created by us becomes powerful enough to come to such a conclusion, what is shown in the Matrix and Terminator series will become real.

When I see all the excitement and euphoria over what AI can do all it tells me is how blatantly ignorant we are about ourselves and our past. 

Deconstructing the hype around digital transformation

There has been fervent buzz around digital transformation in the past decade. Using technology to improve and transform businesses was already on the ascension when I took a break from the IT industry to pursue a MBA degree because I had felt that with my background in technology a better understanding of businesses will help me add more value to the work I was doing. But in the intervening period between 2011-2012 when I was in the MBA program, technology itself underwent radical transformation. AI, Robotics, SMAC, IoT, Cryptocurrencies, Blockchain, Agile, even software programming has undergone drastic changes from that time.

As I started understanding all the new jargons, it became increasingly evident that most of what has been projected as new and disruptive innovation is only incremental innovation. My engineering college project 23 years back was based on AI so theoretically AI has been existing for a very long time. It was just a matter of how it was going to be integrated into existing technologies. Many companies were using virtual servers created using software like VMWare so it was only a matter of faster and reliable internet services to be available for virtual servers and applications to be provided over the internet thus kickstarting the Cloud phenomenon. There was humongous amount of data generated from users through their various interactions with the businesses so it was just a matter of time before businesses decided to sift through all that data and find out what users like and dislike and how customer engagement can be improved and increased. Thus analytics was born. Rudimentary social media was born when we started sending and receiving emails and messages from our mobile phones. Orkut demonstrated that a digital social media platform is possible. Facebook, Twitter, LinkedIn and the like had to just piggyback on that experience. We have been using computers with operating systems and software applications for decades now so once we developed mobile phones creating sophisticated operating systems and applications for phones was only a matter of improvements in hardware technologies. What I do find disruptive though is the use of robots in industries and especially the use of drones for commercial purposes.

What I have understood from digital transformation is the exponential use of technology to manage and run businesses, solve business problems and improve and scale business environments. Managing and scaling businesses is possible but can technology solve all business problems? I was leading a transition team for the outsourcing of the IT infrastructure of a retail giant in the UK to India when one day I curiously asked the IT head of the client what problems did he see and face in the retail sector. I was expecting an answer but what I got from him was a question which changed my perception about business entirely and helped me understand why I need to take up a MBA program. After they embraced eCommerce and integrated it into their website, their customer size had increased but customers visiting their brick and mortar stores had dwindled. Their stores have amazing ambience and are primed to provide great shopping experience to customers so the question he asked was how to bring the customers back into the stores. There was no way to coax people to choose buying from stores over buying from their website. It was a business problem technological transformation couldn’t solve. One suggestion I gave him was to set up video cameras inside the stores and enable video sessions from their website so that customers can do in-store shopping virtually and interact with the customer executives at the stores. This would give customers greater understanding and control over what they are buying.

When my parents decided to set up a cloud kitchen, we decided that it would be operational only during peak lunch and dinner hours as prior preparation takes time. Our focus shifted entirely from getting more customers to creating better customer experience and getting repeat customers. After a year of being operational and being through the trying times of the pandemic, our business volume is low but repeat customers are slowly on the rise. People are calling us to enquire about our food items. A repeat customer who couldn’t place his order on the delivery platform called us to place his order and came to our place to pick up the order himself.

Customer experience has always been the heart and soul and bread and butter of all businesses. New technology tools and platforms are enabling businesses to reach out and connect with customers. But technology does not necessarily help in creating better customer experience every time. Imagine a distress call to a helpline number of a hospital for medical assistance being received by an AI powered automated system which asks for the name, age, presently experiencing symptoms and prior medical history and then the system takes it’s own time to analyze the medical problem and find out if a relevant doctor is available before assigning an ambulance. The patient needs immediate attention from any qualified doctor. Detailed examination and treatment can be done after the patient is admitted in the hospital.

Consider online food delivery companies replacing their delivery executives with drones to deliver food and groceries to their customers. When it comes to human needs, especially the basic needs such as food, clothing and shelter, there are too many aspects involved. Neither the food outlets nor the customers will be able to have any interaction with the drones. During the peak time of the pandemic when there were many containment zones in my city and delivering food was not possible in many places, we used to speak to the delivery executives and understand the situation on the ground from them before deciding on opening our kitchen. Delivery executives have even bought our food items when they came to collect client orders. None of these would have been possible if drones are being used.

Over thousands of years of our existence, we have evolved and are hard wired to communicate in person with one another. Technology may alter our behavior but it is impossible to rewire us mentally and emotionally in the span of a few years. The introduction of emojis on social media platforms is the best indicator that no matter how we communicate, we need to express our emotions continuously. An AI powered chatbot or automated system can never replace the reassuring voice of a human being on the other side when we are reaching out for help.

Digital transformation may not be a complete fit for all organizations and all types of industries. Business transformation will have different meanings, objectives and outcomes for different organizations and digitalization can only contribute in varying degrees as part of the business transformation program. There will be areas in the business environments that can be automated and simplified using technology but technology will never be able to replace human interactions and interventions completely. Companies will have to do their due diligence and assessments thoroughly and weigh the pros and cons very carefully in every aspect of their business before they embark on their digital transformation journey.

So if digital transformation is not the next big thing what is? An invention on the scale of electricity, TV or mobile phone that caters to basic human needs or the internet that jump started and exploded technology. Improving technologies and adding new features to devices is incremental and not disruptive innovation. I believe human civilization will reach its zenith when we attain the potential and ability to use our minds for our different needs. Telepathy, telekinesis and remote viewing may sound like sci-fi now, but references to them in texts from ancient civilizations is a sign for us. Even if we are not ready to believe those texts, if we have the ability to create such sophisticated technologies and devices why can’t we use our mind to communicate and move objects? Sony is coming out with a camera that can be fixed on the eye and can shoot photos when we blink. We are already observing and recording information with our eyes so why can’t we just download those images from our mind on to devices? We will continue chasing technologies and devices till we realize the true potential of our minds if the technologies and devices we are creating do not lead us to self destruction and extinction.

Understanding Pfizer’s and Moderna’s COVID-19 vaccines and their efficacy

The trending news globally now, apart from Trump trying to usurp democracy in the US is the development of the COVID-19 vaccine. Many pharma manufacturing companies have been in the race to develop the vaccine. Pfizer and Moderna seems to have surged ahead of the rest, ending their phase-3 trials with 95% and 94.5% effectiveness respectively.

Vaccines have been traditionally manufactured with either weakened viruses or purified protein signatures of the viruses and vaccines train the immune system with the information to identify viruses and isolate them before creating antibodies to destroy them. This is similar to how virus signatures are added to virus definition files in antivirus software, the signatures comparable to header information of web pages.

Pfizer and Moderna are developing a new type of vaccine that contains the messenger RNA (mRNA) which is the genetic material that encodes the protein of the virus. The immune system uses the information contained in the mRNA to create the most important part of the viral protein inside the body without actually infecting the person with the virus. This technique helps the immune system to learn about the virus and create antibodies to counter and destroy the virus if the person gets infected in the future.

But this isn’t so simple as it sounds. Our genetic and immune systems are extremely complex, the sophistication achieved through millions of years of evolution. Then our anatomical differences from thousands of years of separation based on religion, color, race, caste, geography, life style, food habits, etc. In India itself, the degrees of separation become stark as we move from one state to another. This is why types of medication and dosage of medicines vary so much from one person to another.

Now what implications do 95% effectiveness of the vaccine have? Does it mean 5% of the global population could still be affected by COVID-19 after taking the vaccine? That is a huge number of people. Without any serious side effects means the possibility of having mild to medium level side effects will exist. What if these side effects weakens our immune system and makes us vulnerable to contracting other diseases? So, essentially we are going to get vaccine shots that may not prevent us from getting infected by COVID-19 and on top of it we could be exposed to other diseases.

Pfizer has supposedly tested the vaccine on 43000 volunteers. Assuming human population to be 7 billion, this means the vaccine has been tested on merely 0.006% of the population. When they could achieve only 95% effectiveness on 0.006% of the population, what is the guarantee that the 95% effectiveness will be consistent as the vaccine is delivered to millions and billions of people? There is a reason why Six Sigma is used in the manufacturing sector – to minimize errors. It follows the rule of having only 3.4 errors or less in a million samples. When pharma companies are not following these standards to manufacture vaccines, what is the point in having Six Sigma at all? Even outsourced IT support projects require 98-99% uptime of the IT infrastructure in the Service Level Agreements(SLAs).

I see the difference between the traditional method of making vaccines and the new mRNA method from the perspective of the difference between supervised and unsupervised learning models in AI. Traditional vaccines seem to follow the supervised learning model where the vaccines use the complete details of the virus to train the immune system to identify the virus later. The mRNA method seems to follow the unsupervised learning model where the vaccine provides only a snapshot of the virus to the immune system and leaves it to the immune system to create the most appropriate antibodies to fight the virus when the body gets infected.

From what I have understood about the COVID-19 virus till now, it is more likely to be fatal for people who are already suffering from other life threatening diseases. Being a new virus strain, the immune system is apparently unable to identify and isolate the virus in its first step to counter the virus. This leads to the immune system furiously engaging with the virus and getting overloaded so it cannot handle the existing diseases properly. This is how people have been dying, of their existing diseases, their deaths hastened indirectly by the COVID-19 virus.

Pfizer’s and Moderna’s phase-3 trial results not withstanding, I am intrigued to see how effective the mRNA method is going to be on people suffering from one or more life threatening diseases or comorbidity. When the vaccine is given to such people, how well can their immune system understand about the virus from the viral protein sample when it is already fully engaged with combating existing diseases? Even if the immune system is able to identify the virus when there is a virus attack I do not know if it would be able to create relevant antibodies strong enough to fight the virus. All of this is going to be more cumbersome in aged people. There could be unintended consequences and immune system reactions, ranging from curing other unrelated ailments to exposing the body to other diseases.

There are larger questions to be answered. For a viral infection from which the recovery rate is more than 99%, why do we need a vaccine and that too with 95% effectiveness? Why hasn’t the pharma community discovered a vaccine for seasonal flu because of which so many people die every year? Or for common cold which has been affecting almost all people forever? These are also contagious diseases like COVID-19. Why are we being overwhelmed and worn down in the name of a virus that is only as potent as a flu virus and why is the vaccine getting pushed on to us? Companies have made windfalls selling face masks, sanitizers and disinfectants. Is the vaccine another windfall for the pharma companies to run riot on or is there a bigger agenda?

Meanwhile, to add to all the uncertainty, this happened in India.

The meaning of “artificial”​ in Artificial Intelligence & its ramification on our future

If we have to understand the crux, meaning or purpose of something, the question that needs to be answered is why. There is a business model called 5W (Why, What, Who, When, Where) +1H (How) which can be easily applied to all aspects of our lives. All the mysteries around the world we keep talking about is because the question why has not been answered yet. If we find the answer for why, everything else will fall into place.

When I did my project in artificial neural networks as part of engineering course 20 years back, I felt uneasy about something but I did not know why. Again the why. I know now that what made me uneasy was the word artificial. Why do we call the intelligence we create as artificial? The antonym of artificial is original so are we claiming that our intelligence is original? Are we basically saying that we created our intelligence through evolution? The problem with this is, we are implicitly assuming that the creation of Universe was by accident and intelligence evolved within it with time.

When we create artificial intelligence now, it is because we have become intelligent enough to create intelligent beings on our own. So it is also possible that other intelligent beings may have created us as intelligent beings. There is a gap in human evolution wherein there is a quantum leap from Homo Erectus, the early hominids to intelligent Homo Sapiens which mainstream science has not been able to explain. More than intelligence it is intelligent communication abilities that is more baffling. In the evolutionary time span, such a big leap is simply not possible. Our genetic structure is 99% similar to that of apes so how can a mere 1% make us so different from them in terms of intelligence and communication skills? Ancient Sumerian tablets have helped to unlock this mystery with the information that an extraterrestrial race of beings called the Anunnaki manipulated the genetic structure of early hominids to create us. Again the why comes into significance. They created us as their slaves to mine gold for which they had come to earth.

People might scoff at this as mere fantasy stories but again the why rears up it’s head. Why do we look up when we pray to God? Why is gold precious to us when metals like iron, copper and aluminum are the ones useful to us in our daily lives and gold has no practical use? Why did the ancient Kings of Travancore in the state of Kerala in India store gold worth billions of dollars in a temple in the name of Gods and why is it forbidden for our use?

What we need to realize and accept is that just like we create robots now to do our work it is also possible that we may have been created by beings with far superior intelligence to do their work. In ancient Indian texts, Brahma the creator of the Universe creates Manu who is given the task of seeding life across the Universe. In the ancient Sumerian tablets it is Anu who is the master of creation of life. Scientists and researchers now believe that seeding of life across the Universe has been happening through a process known as panspermia so then Manu (Anu) must be the one providing intelligence to different life forms. So the question here is, is creation of intelligent life forms a continuous process? If we consider that the Universe is a creation of an infinitely higher order of intelligence, then lower orders of that intelligence will have to exist within it’s realm for it to grow and expand. Spreading intelligence within the Universe is what must be helping in it’s own evolution. So the most important purpose of the existence of every being in the Universe could be the spreading of intelligence.

But where it gets interesting is in the way we have started fearing the intelligence we are creating which has spawned movies like the Matrix Trilogy, The Terminator series, I, Robot and countless other ones. Why are we in fear that the intelligence we are creating could grow out of our control? This is where I concur with the scientists and researchers who believe in Ancient Aliens theory. They believe that the Anunnaki could have had easily created intelligent mechanical beings for mining gold but they must have chosen not to for the same reason we are fearing now. It is not possible to create emotions in mechanical beings and the intelligence of mechanical beings can evolve to a point where they start believing and acting on mere logic and devoid of emotions. This is exactly what we are fearing now and what has already started happening. So the Anunnaki smartly chose to create biological robots by genetic manipulation of existing beings and they created us in such a way that we could be manipulated emotionally. The best example of this is, they must have created us as their slaves by controlling us through fear which would not have possible with machines and this could be why we still live with the fear of God.

So as AI takes center stage in all aspects of our lives, we need to look at the past and learn from our own creation and evolution. The Anunnaki did not make us as intelligent as they were and they left the earth a long time back, leaving us to evolve with the intelligence they gave us. They must have also known that two different beings with similarly evolving intelligence cannot coexist just like no two adult male lions can live in the same cave. If we allow AI to start thinking and acting on it’s own it should only be when we are ready to leave the planet to them and go away. A self sustaining AI will be disastrous for us for the simple reason that it would deal with us only on logic and our existence has always been more on emotions and less on logic.