The confluence of the following things will cause machine learning to bring about a revolution:

• Already large amounts of data collected.  Large pools of data have been collected, but haven’t been used as much as they could be.  Machine learning can teach itself about patterns in the data, and then apply these patterns to new data.
• Cloud computing power: per hour computer processing available at low cost. No need to buy computers, just need a device connected to the internet.
• The latest technology available for free.  Core machine learning technology is open source, and the latest versions are publically available at no cost.

This is the first time that the next big technological wave has been so accessible.  All that a person requires is a device connected to the internet and a few dollars to develop world-class scalable machine learning applications.

Moneyball could be real?

Optimising results

The movie Moneyball showed a system whereby using past statistics and raw data, the best training methods could be achieved.

Predicting sports winners

Based on the theory that past behaviour is the best predictor of future behaviour, machine learning will be able to better automatically predict sports winners.  We can see there will be a battle of the algorithms.
Machine learning predicts World Cup winner

Make sports more entertaining?

Formula 1 Uses Machine Learning To Deliver In-Race Predictions To Fans
Using the past to predict the present, machine learning will take in all the measurable factors in a sports game, and then try to predict what events will happen next.  Will this make for an enhanced spectator experience, or not?
 

Supply chain
One of the first ways that machine learning was demonstrated effective, was when machine learning was proven to be more effective at identifying malignant melanomas than top medical experts.
An interesting article on identifying problems in a supply chain takes this further.  Machine learning could be trained in identifying and faulty products in a supply chain, making quality control much more efficient.
10 Ways Machine Learning Is Revolutionizing Supply Chain Management
Crime
There are too many CCTV security cameras for human eyes to actively monitor.  Footage is recorded and then played back later if an incident is reported.
Machine learning technology, having being trained with past footage, will enable machine learning to detect where there is abnormal or erroneous behaviour than what is normal for that particular environment.  The computer algorithm would then draw attention to what is going on, making the cameras a tool for disrupting a crime incident taking place, rather than just being an evidence gathering tool after an incident has taken place.
To extrapolate this to the nth degree, pre-crime may be possible by detecting suspicious behaviour.  Just as machine learning currently can detect melanomas better than human experts, machine learning in the future may detect changes in environments that seemingly foresee crimes about to happen.

Due to government legislation, not many large databases are linked or joined. As data is used more together, it exponentially becomes more powerful and invades privacy.  The pieces of a person’s life can be pieced together and the puzzle becomes clearer.
Did you know that if all your data was linked, I could:

• Tell you what day and time you are likely to go to the supermarket;
• Tell you what you are likely to buy at the supermarket;
• guess how many people and what ages are in your family;
• guess how often you travel; and
• guess where you normally like to travel.

Data on purchases is insightful when properly analysed, and is one of the main data trails that people leave behind them. Over the coming years, retailers will not accept cash as payment, meaning that all payments are traceable.  Bitcoin is not yet a mainstream method of paying, but probably more anonymous and mainstream methods will spring up as people want to protect some of their privacy.

Every day, you present something to strangers that is remarkably unique and that can be used to identify you.  Your face.

Ramifications of the new technology

The confluence of increased use of security cameras, and more effective facial recognition technology will have huge ramifications.
And for most of us, it’s too late to opt out. Any photo you have deleted from publically visible pages on social media has already been saved by internet scanners such as archive.org.
Never again will the public have an assumption of privacy.  Artificially intelligent “bots” will be able to accurately detect a person’s face using facial recognition biometrics that measures different aspects of a person’s face, and matching this to a database of “facial fingerprints”.  Machine learning technology makes it much easier for the bots to guess whose face it is (it doesn’t have to be an exact match), and do it with startling accuracy.  An example of this is the FindFace application, which allows you to you take a photo of a stranger and automatically find their social media profile.
 

Legislation will only be partially effective

Legislation may control the use of facial recognition technology by public and corporate institutions, but the highly effective technology will inevitably be used in unregulated ways.
Publicly posting a selfie of your face is the equivalent of posting a high-resolution image of your fingerprints.  A person’s facial characteristics can be made into a computer file that will serve as a digital fingerprint.  In the future, the entire world’s population “faceprints” will be able to be carried around on a USB stick.
Imagine if a government biometric database was hacked – and the biometric information, or even just the photos, was then distributed on WikiLeaks.  Any image of your face could be used to identify you – and fast technology will link this to all your public information online.

Privacy diminished

Government agencies will start to share these face databases.  All it takes is for one of these databases to be leaked to the public, and with a simple phone app, you will be able to match a photo to an identity.
Imagine a future where people regularly wear face masks in public to protect themselves from this facial recognition.  Doing so may become normal behaviour.  In future, government legislation may prevent people from covering their faces under the auspice of anti-terrorism laws, just like the current push against encrypted messaging.
References:
Privacy concerns voiced over photo database link to real-time surveillance
Does Facial Recognition Technology Mean the End of Privacy?
Let’s face it, we’ll be no safer with a national facial recognition database
We Built A Powerful Amazon Facial Recognition Tool For Under $10
DNA facial prediction could make protecting your privacy more difficult