To some it’s scary and to others it’s exciting, but the scientific consensus is that artificial intelligence will have a drastic impact on humanity, probably within our lifetimes.

What exactly that will look like is a debated range of sci-fi scenarios. Stephen Hawking made waves when he claimed that the development of AI could end the human race. While real-life Tony Stark Elon Musk and Microsoft’s Bill Gates chimed in with concerns of their own, other experts believe the threat of artificial intelligence has been exaggerated. What everyone can agree on is that developing intelligence is something we should be very, very careful with.

When Google purchased DeepMind to the tune £400 million, the London-based AI startup had some pretty firm ground-rules regarding the two companies’ relationship. Demis Hassabis, theDeepMind CEO, said that a condition to their acquisition by Google was for Google to form an internal ethics committee. DeepMind also refuses to allow any of their technology to be used for weapons or military interests.

Hassabis has announced that he and many of the top minds currently working with AI research will be meeting in New York in early 2016 to discuss and debate ethical issues surrounding their work. Although no official list of participants has been released, big players such as Apple and Facebook will almost certainly have representatives present.

Since purchasing the AI company, Google has been using DeepMind’s technology in a wide array of implementations. Artificial intelligence has improved Google’s image recognition technology and is also helping services like Google Now anticipate user’s needs more accurately. Talks like the one expected to occur in New York will likely serve to create ethical frameworks that will guide the development of this and other technologies?

Given that we live in a world dominated by capitalistic desires, can humanity entrust the synthesis of such a pivotal base – the AI ethical framework, in the hands of Google, Facebook and Apple? Do these companies not have a vested business (profitable) interest in AI that may bias their respective inputs towards individual gain over that of humanity?

It maybe is time for humanity to be more involved in what will define the next evolutionary cycle. After all its seems that our position in the food chain is under threat!

I have a confession to make: I’m addicted to trip planning.

I blame my first-World problem on two major reasons: curiosity and dynamic pricing. While curiosity might fuel my craving to explore new places, dynamic pricing has taken this craving to a whole new level.

Image source

Long gone are the days I would walk into a travel agency and accept what was given. Today, I spend days monitoring flight prices online – always in the outlook of a good deal.

Early 2013, travel search engine Kayak introduced a fare-forecasting tool that allows travelers to assess whether the search prices will rise or fall within the next week. This was the first time I began to wonder how the travel industry could forecast demand and optimize its prices. What lies behind those flight purchase recommendations?

I soon learned that Kayak uses large sets of historical data from prior search queries and complex mathematical models to develop its forecasting algorithm. I imagined that some of the input variables include the current number of unsold seats, the date and time of the booking (specially the number of days left until departure) and the current competition on the same route. But what if Kayak also learned more about my personal preferences based on my past behavior? If there were only one window seat left on that flight, would it change its purchase recommendation from wait to buy? Would it be able to recommend me a trip, knowing I prefer nature over cityscapes, am most likely to buy on Tuesdays and usually travel over the weekend?

Image source

A lot has been done on dynamic pricing since 2013, of course. One of the latest companies to join the trend is AirBnB. Its hottest new feature, Price Tips, helps hosts to easily price their listings dynamically by using the company’s new open source Aersosolve machine learning tools. Their use of machine learning isn’t just limited to the standard dynamic pricing, though: their models automatically generate local neighbourhoods, rank images and learn about AirBnB hosts’ preferences for accommodation requests based on their past behaviour.

Image source

What does this mean for trip planning addicts?

Well, I'm personally looking forward to the time when choosing my next travel destination is made simpler. I picture personalised travel discovery and contextual insights - the kind that take into account both, your conscious and unconscious preferences. The Texas-based startup, WayBlazer, already is tipping into this by employing IBM Watson cognitive computing technology to listen and understand its customers and present considered, tailored travel suggestions.

What could be next?

picture: www.hitc.com

Machines informing financial positions and investments. Are we simply moving forward with new technology or we are just down right lazy?

While our past is cast in stone and unchangeable and our present is an ongoing phenomenon; our future however is more uncertain. Nowhere else are the consequences of these uncertainties more significant than the world of investment banking. Within the investment banking area, particularly in the light of artificial intelligence (machine learning), there are two critical Investment banking products to consider: Mergers and acquisitions (M&As) and equity stock markets (shares).

Let us consider M&As as they were done previously, companies (medium and large scale) required the services of industry experts to identify the best fit M&A companies and provide a web of information needed for the closure of deals. With shares in the past, the biggest challenge for stock traders was predicting stock prices and market trends in the financial equity market to maximize returns. One of Warren Buffet’s more inspiring quotes, “It’s easier to look back than to look into the future’’, emphasizes the difficulty in predicting stock behavior. In those days, stock trades involved buying stocks on a physical trading floor and required a network of persons to execute a trade, with traders relying more on bare intuition and basic trend analytic tools. These all happened in the analog world!!

In the digital age of today, there are different algorithms and indicators that have been developed to change interaction between buyers and sellers. Buyers can find a best fit company in a matter of hours, saving them the cost of research and precious time. The use of artificial intelligence tools such as machine learning and big data have emerged with several programs.  These programs obtain as much available historical data about a company and seek to create a relationship between this historical data and future prices of the shares of the company. To achieve this, investors and researchers create different algorithms (decision trees, support vector machines, Naive Bayes classifiers, etc.) and price indicators. Some of these algorithms have been designed to be so intelligent that they are not rigid to any single investment approach but adapt to market trends and situations.

All this is not to say that traditional investment banking as was practiced in the ‘analog world’ is dead. In fact, very intelligent and experienced bankers and traders are still very relevant in the process of negotiating and structuring deals to reduce tax payments and place stock trades based on knowledge of future occurrences. A valid question though is: “How much longer will human experience still be required?’ Currently, a tech start-up based in Cambridge, Massachusetts, USA called Kensho has received a lot of buzz and huge investments from large corporations such as Goldman Sachs, Google ventures and Consumer News and Business Channel (CNBC) for its development of a software called Warren. Kensho believes that Warren will replace financial analysts in investments transactions. The company boasts of the capacity of Warren to search financial data and reports and provide replies in natural language within seconds.

I have a friend, George, back home in Nigeria. He is one of the hundreds of thousands of investment bankers in the world. Does this mean he is going to be jobless in the near future? I know machines are smart or can be programmed to be smart, but we all realize the intuition of humans and the capacity of mankind to reason through complex and confusing situations. In such situations, simple algorithms may fail, especially the situation that has not obeyed the rules. Take for instance, the earthquake and Tsunami in Japan in 2011 (deemed the costliest natural disaster in history). Then you need a human, then you need someone like my friend George. Now, I am not in anyway suggesting that technology is not the way forward. However, in going forward technologically, we need to allow for man to move machine and machine to listen to man. After all, even unmanned space shuttles are still “manned” by men on Earth.

References

Walters, Richard 2015 Investor rush to artificial intelligence is real deal http://www.ft.com/intl/cms/s/2/019b3702-92a2-11e4-a1fd-00144feabdc0.html#axzz3oG7w4n00 Financial Times 4^th January 2015

The Economist 2015 Artificial intelligence: Rise of the machines

http://www.economist.com/node/21650526/print1/13Artificialintelligence 

The Economist 5^th August 2015

Kensho Press https://kensho.com/#/press

Have you ever wondered how to world looks through the eyes of a party raver having a Psychedelic experience? Well Google may have just made this ‘legally’ possible (and yes without one having to take any illegal substance) through one of their new platforms.

A few months ago, Google had announced that they now possess the technology to gaze inside the mind of an artificial intelligence program. Having invested heavily in machine learning technology - Google is one of the world’s biggest backers of artificial intelligence development. Googles recent acquisition of a British company ‘DeepMind’ is a testament of Googles vigor to unlock the potential of Artificial intelligence and it is through the Deep dream platform that one can perceive what a machine ‘Sees’ or ‘Dreams’

Google Image

The network uses 10-30 staked layers of artificial neurons with each layer adding incrementally to the results of its predecessor in order to obtain the final answer as produced by the last layer.  On the lines of image recognition, the network seems to set a new benchmark by returning results better than anything before and as a by-product, it can also “dream.” These artificial dreams output some captivating images to say the least, going from virtually white noise to something that looks out of a surrealist painting or probably the vision of our raver above having a psychedelic trip - and you thought Machines can’t be creative!

To access image patterns of how Google’s neural network “sees” or “dreams” to go through this post 

The above seems really creative and is all well however the million dollar question remains – how dependable is Artificial Intelligence? Where do humans draw a line in the sand regarding machine driven vs human driven output. A stark reminder of the current AI technological limitations were made evident to Google the hard way. Google Photos employs advanced artificial neural networks to analyse gazillion images, interpret them and return the right one that a user has queried using the google search engine. The app uses face and object recognition software to automatically tag and sort photos however in a recent instance had mistakenly tagged pictures of a black couple as ‘Gorillas’. Google had to issue an apology after Jacky Alcine, the man in the picture, was outraged to see the racially charged term appear in the app. Alcine also tweeted a screenshot showing every image of his friend was being tagged, and suggested the reference images Google had collected did not have black people in mind.

The above incident confirms one of our worst fears that artificial intelligence is racist, or so it seems. The supposedly dumb and stupid data-in and data-out machines, that strived to always catch up to us humans have acquired prejudice.

When I first heard about Machine learning I was really pissed off because of the below three reasons

1. Machine learning, F*** yet another technology to make life complicated.
2. I should again update my small data base (brain) with this new technology and should override the technologies this might erase. Hectic job L…!!!
3. Can someone write program to keep human mind updated automatically with the new technologies inventing all the time. I hope this happens soon.

With this notion I started reading about machine learning and to my surprise it’s actually fascinating and interesting than I ever thought. Machine learning is fascinating because you cannot achieve the capabilities and results seen by machine learning methods with any other methods. The most fascinating about machine learning is you make programs to learn from the data, examples and experiences and that programs make programs to address different problems.

Curious to know more about machine learning I started researching about the usage of machine learning in different fields. I found that machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Machine learning is so pervasive today that you probably use it dozens of times a day without knowing it.

After all this research the next question that struck me was how can I make money with machine learning???

I was reading an article in FORBES which points that 80-90% of the startups fail every year. This struck me with an idea on why cannot we use machine learning to predict the success of the startups??

I guess with some effort it is possible to gather data from these ventures and also from the other successful ventures which could actually be the data set for the machine learning methods. Data about thousands of companies can also be collected from online sources such as CrunchBase etc.

At this point of time I have no idea about the complexity of addressing this challenge and the number of variables that need to be considered to get more accurate results but I am sure machine learning can definitely be useful in this regard to determine the exit likelihood of private companies and explain why they might succeed or fail by highlighting their strengths and weaknesses. This analysis can also be helpful to suggest ways to improve the distressed startups.

The next question that bothered me was, do any venture capitalists use machine learning to predict startup success?

For Venture Capitalists, predicting, measuring and evaluating the success of the startups they invest in is risky business. As per my knowledge (I admit my knowledge is limited in this area), Very few VCs actually use software with machine learning techniques to predict success. I believe the reasons as below

1. VC’s sample is too small compared to the market, so it's difficult to allocate money to build a piece of software to process small sample data.
2. Even if VC’s are planning to expand their databases with other VC portfolios, they wouldn't have enough accurate data and insights about those companies.
3. Humans change plans. People is the most invariable factor in every single company. It's also the most important one because people have an impact on everything and everything can have an impact on the people. I believe investors almost always invest in the right combination of management team and product and not just only product. In this context no software can accurately predict the human behavior and this is a major bottleneck for creating such software.

VC’s at the moment use tools like Mattermark in order to detect and track startups, sectors and market trends and try to bid on the winner of each category.

I believe a software with machine learning technique would address all above problems and provide VC’s with more accurate results than they would get from other sources.

As per my knowledge very less work is being done in this area, I suppose only one startup (Trenify.io) is actually working on implementing this but I suppose their accuracy is also based on certain limitations/assumptions on human behavior.

I believe Machine learning to predict the startup success could be one area for people planning to launch new ventures.

I would love to see people commenting to this post and discussing if this is a viable option for a startup?? Will this startup idea be successful before it predicts the future of other startups???