Introduction to AI

AI enables us to build amazing software that can improve health care, enable people to
overcome physical disadvantages, empower smart infrastructure, create incredible
entertainment experiences, and even save the planet!

Watch the following video to see some ways that AI can be used.

Artificial Intelligence isn't coming, it's here. It's the defining technology
of our time. But it's human ingenuity that makes it so powerful. Jane started
losing her vision when she was 12. She uses Seeing AI. I crept down the stairs
and out the door. I crept down the stairs and out the door. Which enables her
to read and navigate her world.

Microsoft AI is helping farmers adapt to climate change. We can use technology

to help feed the world without wrecking the planet. Enabling wildlife
conservation. The ability to use AI really democratizes science.

Wild Me uses images from any source to help study endangered species. Creating
new ways for people with hearing loss to communicate. I want the exact same
information that my hearing friends have. It's more than just imitating
intelligence or sounding like a human. Microsoft AI is changing every industry
from health care.
Now you can see these finite details,the relationship with many structures. To
engineering and telecommunications. We are using Artificial Intelligence to
change the way we are relating to customers. AI lets us put people first.
The Microsoft tools really enabled us to breathe digital life into an iconic
brand persona. The product that we make is not a thing in a box. Microsoft AI
empowers others to create experiences and potential. This is just the
beginning.

Artificial Intelligence, it's just opened up this whole world. The future we
invent is a choice we make.

What is AI?

Simply put, AI is the creation of software that imitates human behaviors and capabilities.
Key workloads include:

 Machine learning - This is often the foundation for an AI system, and is the way we
"teach" a computer model to make prediction and draw conclusions from data.
 Anomaly detection - The capability to automatically detect errors or unusual activity in a
system.
 Computer vision - The capability of software to interpret the world visually through
cameras, video, and images.
 Natural language processing - The capability for a computer to interpret written or
spoken language, and respond in kind.
  Knowledge mining - The capability to extract information from large volumes of often
unstructured data to create a searchable knowledge store.

Understand machine learning

Machine Learning is the foundation for most AI solutions.

Let's start by looking at a real-world example of how machine learning can be used to
solve a difficult problem.

Sustainable farming techniques are essential to maximize food production while

protecting a fragile environment. The Yield, an agricultural technology company based
in Australia, uses sensors, data and machine learning to help farmers make informed
decisions related to weather, soil and plant conditions.

View the following video to learn more.

Video:
Right now, artificial intelligence helps us meet the needs of today, so we're
prepared for tomorrow.
By 2050, we need to produce 60 percent more food.

So how are we going to feed the world without breaking the planet?

Using Microsoft Artificial Intelligence, we can reduce waste and produce more
food.

Any grower will tell you every row,every crop is different.

We can use Microsoft AI to make local predictions about light, wind, rain.

This helps farmers know when to plant, irrigate, and harvest.

It's making a difference. Artificial intelligence helps farmers grow more

while wasting less.

You can find out more about how the Yield is using machine learning to feed the world
without wrecking the planet here.
How machine learning works

So how do machines learn?

The answer is, from data. In today's world, we create huge volumes of data as we go
about our everyday lives. From the text messages, emails, and social media posts we
send to the photographs and videos we take on our phones, we generate massive
amounts of information. More data still is created by millions of sensors in our homes,
cars, cities, public transport infrastructure, and factories.

Data scientists can use all of that data to train machine learning models that can make
predictions and inferences based on the relationships they find in the data.

For example, suppose an environmental conservation organization wants volunteers to

identify and catalog different species of wildflower using a phone app. The following
animation shows how machine learning can be used to enable this scenario.
 1. A team of botanists and scientists collect data on wildflower samples.
2. The team labels the samples with the correct species.
3. The labeled data is processed using an algorithm that finds relationships between the
features of the samples and the labeled species.
4. The results of the algorithm are encapsulated in a model.
5. When new samples are found by volunteers, the model can identify the correct species
label.

Machine learning in Microsoft Azure

Microsoft Azure provides the Azure Machine Learning service - a cloud-based platform
for creating, managing, and publishing machine learning models. Azure Machine
Learning provides the following features and capabilities:
Understand anomaly detection

Imagine you're creating a software system to monitor credit card transactions and
detect unusual usage patterns that might indicate fraud. Or an application that tracks
activity in an automated production line and identifies failures. Or a racing car telemetry
system that uses sensors to proactively warn engineers about potential mechanical
failures before they happen.

These kinds of scenario can be addressed by using anomaly detection - a machine

learning based technique that analyzes data over time and identifies unusual changes.

Let's explore how anomaly detection might help in the racing car scenario.

1. Sensors in the car collect telemetry, such as engine revolutions, brake temperature,
and so on.
2. An anomaly detection model is trained to understand expected fluctuations in the
telemetry measurements over time.
3. If a measurement occurs outside of the normal expected range, the model reports
an anomaly that can be used to alert the race engineer to call the driver in for a pit
stop to fix the issue before it forces retirement from the race.

Anomaly detection in Microsoft Azure

In Microsoft Azure, the Anomaly Detector service provides an application

programming interface (API) that developers can use to create anomaly detection
solutions.

To learn more, view the Anomaly Detector service web site

Understand computer vision

Computer Vision is an area of AI that deals with visual processing. Let's explore some of
the possibilities that computer vision brings.
The Seeing AI app is a great example of the power of computer vision. Designed for the
blind and low vision community, the Seeing AI app harnesses the power of AI to open
up the visual world and describe nearby people, text and objects.

View the following video to learn more about Seeing AI.

Seeing AI is a Microsoft research project for people with visual impairments.

The app narrates the world around you by turning the visual world into an
audible experience. Point your phone's camera,select a channel, and hear a
description. The app recognizes saved friends. Jenny, near top right, three
feet away. Describes the people around you, including their emotion. 28-year-
old female wearing glasses looking happy. It reads text out loud as it comes
into view, like on an envelope. Kim Lawrence PO box. Or room entrance.
Conference 2005. Or scan and read documents like books and letters. The app
will guide you and recognize the text with its formatting. Top and left edge
is not visible, hold steady. Lease Agreement. When paying with cash,the app
identifies currency bills. $20. When looking for something in your pantry or
at the store, use the barcode scanner with audio cues to help you find what
you want. Campbell's Tomato Soup. When available, hear additional product
details. Heat in microwave bowl on high. Even here descriptions of images in
other apps like Twitter by importing them into Seeing AI. A close up of Bill
Gates. Finally, our experimental features, like scene descriptions to get a
glimpse of the future.
I think gets a young girl throwing a frisbee in the park. Experience the world
around here with the Seeing AI app from Microsoft.

Computer Vision models and capabilities

Most computer vision solutions are based on machine learning models that can be
applied to visual input from cameras, videos, or images. The following table describes
common computer vision tasks.
Computer vision services in Microsoft Azure

Microsoft Azure provides the following cognitive services to help you create computer
vision solutions:

Try this

To see an example of a how computer vision can be used to analyze images, follow
these steps:

1. Open another browser tab and go to https://aidemos.microsoft.com/computer-vision.

2. Use the demo interface to try each of the steps. For each step, you can select images and
review the information returned by the Computer Vision service.

Understand natural language processing

Natural language processing (NLP) is the area of AI that deals with creating software
that understands written and spoken language.

NLP enables you to create software that can:

 Analyze and interpret text in documents, email messages, and other sources.
 Interpret spoken language, and synthesize speech responses.
 Automatically translate spoken or written phrases between languages.
 Interpret commands and determine appropriate actions.

For example, Starship Commander, is a virtual reality (VR) game from Human
Interact, that takes place in a science fiction world. The game uses natural language
processing to enable players to control the narrative and interact with in-game
characters and starship systems.

Watch the following video to learn more.

So think about this. How often are you alive when the dawn of a new medium
comes out?

I knew that I had to be a part of VR because there are no rules for what VR
should be. Starship Commander is a choose your own adventure VR narrative that
uses natural language processing, so you can actually have a real conversation
with the characters.
I'm Sergeant Sarah Pearson. I've been assigned to your wing today. When we
started making Starship Commander,we had to figure out how do we get
technology that can understand what a person is saying so that we can have
these new narrative experiences or science-fiction game,we come up with new
words.

No off-the-shelf software could let us put in customers.Who are the Ecknians?

In our story, the bad guys are called the Ecknians. Every time we said
Ecknian, you would come up with really weird words that were similar to other
things that you would say, and so we would get confused all the time. So we
needed to make sure that we could supply our own script. We started using
Microsoft Cognitive Services to power the back-end of Starship Commander.
Well, it's cool.
Once we plug custom speech service into Starship Commander, I had tears in my
eyes when I was playing it because it's like, oh my God, this is it.
This is the experience we want. Open the hatch. Voice print accepted. Welcome,
commander. You can essentially put your script in there and it knows,"Oh,
these are the ways that people are going to talk, and these are the phrases
that they're going to say." Cool. What's my mission? Must be something pretty
important.
Then the second thing too was the intent detection. With LUIS, you basically
give it a couple of statements. After you've given it a couple of statements,
you crunch it,and you start typing in your own statements,and it's shockingly
scary how good it picks up,even stuff that you didn't put it in there.
Computer, use autopilot,and let's be on our way. Autopilot engaging.
I've never seen an opportunity like this,where there's a wide open playing
field. No one has written the rules yet.
There's going to be tons of amazing software out there,and I just can't wait
to see what other people write.

Natural language processing in Microsoft Azure

In Microsoft Azure, you can use the following cognitive services to build natural
language processing solutions:
Try this

To see an example of a how you can use natural language to interact with an AI system,
follow these steps:

1. Open another browser tab and go to https://aidemos.microsoft.com/luis/demo.

2. Use the demo interface to control the lighting in the virtual home. You can type
instructions, use the microphone button to speak commands, or select any of the
suggested phrases to see how the system responds.

Understand knowledge mining

Knowledge mining is the term used to describe solutions that involve extracting
information from large volumes of often unstructured data to create a searchable
knowledge store.

Knowledge mining in Microsoft Azure

One of these knowledge mining solutions is Azure Cognitive Search, a private,

enterprise, search solution that has tools for building indexes. The indexes can then be
used for internal only use, or to enable searchable content on public facing internet
assets.

Azure Cognitive Search can utilize the built-in AI capabilities of Azure Cognitive Services
such as image processing, content extraction, and natural language processing to
perform knowledge mining of documents. The product's AI capabilities makes it
possible to index previously unsearchable documents and to extract and surface insights
from large amounts of data quickly.

Challenges and risks with AI

Artificial Intelligence is a powerful tool that can be used to greatly benefit the world.
However, like any tool, it must be used responsibly.

The following table shows some of the potential challenges and risks facing an AI
application developer.

Understand responsible AI

At Microsoft, AI software development is guided by a set of six principles, designed to

ensure that AI applications provide amazing solutions to difficult problems without any
unintended negative consequences.
Fairness

AI systems should treat all people fairly. For example, suppose you create a machine
learning model to support a loan approval application for a bank. The model should
make predictions of whether or not the loan should be approved without incorporating
any bias based on gender, ethnicity, or other factors that might result in an unfair
advantage or disadvantage to specific groups of applicants.

Azure Machine Learning includes the capability to interpret models and quantify the
extent to which each feature of the data influences the model's prediction. This
capability helps data scientists and developers identify and mitigate bias in the model.

For more details about considerations for fairness, watch the following video.

Video:
So right now, we do live in a society that is unfair and biased in many
different ways. I think the whole point of focusing on fairness in AI systems
is to make sure that the systems that we develop and we deploy reduce
unfairness in our society rather than keep things at the same level or even
make it worse and for me that's really the goal. We've seen a lot of new
stories over the past couple years about AI systems that are used to allocate
or withhold opportunities, resources or information in domains like criminal
justice, employment and hiring, finance, credit, AI systems can reinforce
existing societal stereotypes.

Another example would be cultural denigration, we also have over and

underrepresentation, there isn't a single definition that we can easily
quantify and just integrate into our systems. Fairness relates to not just the
system, the technical component, but it relates to the societal context in
which the system is deployed.

That means that fairness in the context of AI systems is a fundamentally

socio-technical challenge, this means that we've got to have a greater
diversity of people developing and deploying AI systems. What we see is that
the assumptions
and decisions made by teams at every stage of the AI development and
deployment life cycle can introduce biases and that's why this is such an
important topic.
This isn't something that we can just delegate to one or two people and call
it quits and move on, no this is something that everybody has to be thinking
about really actively.

Reliability and safety

AI systems should perform reliably and safely. For example, consider an AI-based
software system for an autonomous vehicle; or a machine learning model that
diagnoses patient symptoms and recommends prescriptions. Unreliability in these kinds
of systems can result in substantial risk to human life.

AI-based software application development must be subjected to rigorous testing and

deployment management processes to ensure that they work as expected before
release.

For more information about considerations for reliability and safety, watch the following
video.

Video:
Reliability and safety is a concern for every AI system we develop. We need to
make sure that the systems we're developing are consistent with the design
ideas we have, and working in the way that it is consistent with our values
and principles. This requires that our systems, our models, are not creating
harm in the world. If there are situations where they may be making mistakes,
we actually push products out there with quantified and well-understood risks
and harms in a way that we actually share that with our users. This is the
guiding principle we have with respect to reliability and safety, and it is
really a concept that applies to every AI product we have in the company. When
we think about safety, the first examples that come to mind are self-driving
cars, but is not even limited to those physical systems, physical agents. We
actually worry about harm to human lives when a machine learning model is
making predictions about people's health in hospitals, when they are making
predictions about diagnosis, wrong systems can lead to harms for people. So
those are the cases that we really worry about because the threat is to human
lives, but this doesn't mean that this issue is only for those physical
systems. Reliability is a big concern and small mistakes may pile up when a
system gets used many times across large group of people, and that's why it is
a concern for everything that we build.

Privacy and security

AI systems should be secure and respect privacy. The machine learning models on which
AI systems are based rely on large volumes of data, which may contain personal details
that must be kept private. Even after the models are trained and the system is in
production, it uses new data to make predictions or take action that may be subject to
privacy or security concerns.

For more details about considerations for privacy and security, watch the following
video.

Video:
Privacy is a fundamental right, and Microsoft has a longstanding commitment to
privacy and security in the systems and products that we build. With AI and
machine learning, we add new complexity to those systems and an increased
reliance on using data to develop those systems, to train the systems. That
increasing reliability and data adds new requirements for keeping the system
secure. For example, we're using more data. We need to ensure the protection
of that data that it's not leaked or disclosed. One of the ways we approach
that is to not remove the data from a customer's device, to run the models
locally on the device, eliminating that potential vulnerability. When we
consider the security aspects of these AI systems, you have to think through
where and how the data is coming from, if it's user-submitted data or a public
data source that's being used in the prediction, how do you prevent that from
being corrupted, having things like anomaly detection or other systems for
detecting
changes in the data that might indicate that there's an adversary trying to
influence the results of the system.

Inclusiveness

AI systems should empower everyone and engage people. AI should bring benefits to all
parts of society, regardless of physical ability, gender, sexual orientation, ethnicity, or
other factors.

For more details about considerations for inclusiveness, watch the following video.

Video:
If we think about what we're really trying to achieve, we want to make sure
that we, as Microsoft, empower and engage the communities around the world.
Making sure that no one is left out. Making sure that we are intentionally
inclusive and intentionally diverse with the approaches that we take towards
AI. We want to make sure that that full spectrum of communities are covered.
If we've truly think about how we can design for the three percent, we can
then solve for the 97 percent at the same time. We want to make sure that
experiences like seeing AI, speech-to-text translation out of PowerPoint, and
other technological examples that we have that we're building upon including
the Microsoft Accessibility Standards represent just an example of the
extensive toolkits that we have here to ensure everyone truly designs for an
inclusive AI future.
As a woman of trans and non-binary experience and also being Latinx as part of
the under represented minority community, ensuring that these future
experiences are truly representative of the under-served minority communities.
We need to make sure that they're involved both in their earliest concepts and
future design and planning, but making sure that they're part of our testing
that we are also building with side-by-side individuals to make sure that we
are not taking an ableist perspective for the population we intend to serve.

Transparency

AI systems should be understandable. Users should be made fully aware of the purpose
of the system, how it works, and what limitations may be expected.
For more details about considerations for transparency, watch the following video.

Video:
Transparency and intelligibility can help us achieve a really diverse range of
goals. So things from mitigating unfairness in machine learning systems to
helping developers debug their AI systems, to gaining more trust from our
users.
There are two sides to transparency; in part transparency means that the
people who create AI systems should be open about how and why they're using
AI, and also open about the limitations of their systems. Transparency also
means that people should be able to understand the behavior of AI systems,
this is what you often hear referred to as interpretability or
intelligibility.

Accountability

People should be accountable for AI systems. Designers and developers of AI-based

solutions should work within a framework of governance and organizational principles
that ensure the solution meets ethical and legal standards that are clearly defined.

For more details about considerations for accountability, watch the following video.

Video:
Despite all the complexity with these new models and with new technology that
can be somewhat unpredictable and somewhat hard to interpret, we are still
accountable for how our technology impacts the world. Second, I think about
accountability as the structure that we put in place so that we make sure that
we consistently are enacting our principles and that we're taking them into
account in everything that we do. Also, part of our accountability is to help
our customers and partners be accountable. We have a set of principles that
guide how we develop and how we sell, and how we advocate for regulation on
facial recognition. Because we feel like all three of those pieces are really
critical to be accountable. We think it has a lot of great uses. We also think
there are a lot of uses that could interfere with people's civil liberties or
pushed aside in a direction that we're really not interested in supporting. So
in moving from our facial recognition principles to practices and processes
that we can follow across the company, what we've had to do is to take a high
level principle, such as our commitment to fairness, and think about how do we
have our development teams view every stage of the development life-cycle with
those principles.
So today we have a set of guidelines that they can follow and think about
these considerations at every step of the life-cycle. That's really a first in
this company, and I think in many other places outside this company as well.

he principles of responsible AI can help you understand some of the challenges facing
developers as they try to create ethical AI solutions.
Further resources

For more resources to help you put the responsible AI principles into practice,
see https://www.microsoft.com/ai/responsible-ai-resources.

Artificial Intelligence enables the creation of powerful solutions to many kinds of

problems. AI systems can exhibit human characteristics to analyze the world around
them, make predictions or inferences, and act on them in ways that we could only
imagine a short time ago.

With this power, comes responsibility. As developers of AI solutions, we must apply

principles that ensure that everyone benefits from AI without disadvantaging any
individual or section of society.