Stay ConnectedWhat is Artificial Intelligence? It’s a big part of our daily lives and you want to know. You need to know. But the explanations are so long and boring. Let me give you something short and sweet.
Join me, Dr. Peper, for 5 minute, pleasing, and easy to understand flash talks about everything artificial intelligence. Short and Sweet AI.
The ethics surrounding AI are complicated yet fascinating to discuss. One issue that sits front and center is AI bias, but what is it?
AI is based on algorithms, fed by data and experiences. The problem is when that data is incorrect, biased or based on stereotypes. Unfortunately, this means that machines, just like humans, are guided by potentially biased information.
This means that your daily threat from AI is not from the machines themselves, but their bias. In this episode of Short and Sweet AI, I talk about this further and discuss a very serious problem: artificial intelligence bias.
In this episode, find out:
Important Links & Mentions:
Resources:
Episode Transcript:
Today I’m talking about a very serious problem: artificial intelligence bias.
AI Ethics
The ethics of AI are complicated. Every time I go to review this area, I’m dazed by all the issues. There are groups in the AI community who wrestle with robot ethics, the threat to human dignity, transparency ethics, self-driving car liability, AI accountability, the ethics of weaponizing AI, machine ethics, and even the existential risk from superintelligence. But of all these hidden terrors, one is front and center. Artificial intelligence bias. What is it?
Machines Built with Bias
AI is based on algorithms in the form of computer software. Algorithms power computers to make decisions through something called machine learning. Machine learning algorithms are all around us. They supply the Netflix suggestions we receive, the posts appearing at the top of our social media feeds, they drive the results of our google searches. Algorithms are fed on data. If you want to teach a machine to recognize a cat, you feed the algorithm thousands of cat images until it can recognize a cat better than you can.
The problem is machine learning algorithms are used to make decisions in our daily lives that can have extreme consequences. A computer program may help police decide where to send resources, or who’s approved for a mortgage, who’s accepted to a university or who gets the job.
More and more experts in the field are sounding the alarm. Machines, just like humans, are guided by data and experience. If the data or experience is mistaken or based on stereotypes, a biased decision is made, whether it’s a machine or a human.
Types of AI Bias
There are 3 main types of bias in artificial intelligence: interaction bias, latent bias, and selection bias.
Microsoft’s Failed Chatbot
Interaction bias arises from the users who are driving the interaction and their biases. A clear example was Microsoft’s Twitter based chatbot called Tay. Tay was designed to learn from its interactions with users. Unfortunately, the user community on Twitter repeatedly tweeted offensive statements at Tay and Tay used those statements to train itself. As a result, Tay’s responses became racist and misogynistic and had to be shut down after 24 hours.
Amazon’s Recruiting Bias
Latent bias is when an algorithm may incorrectly identify something based on historical data or because of an existing stereotype. A well-known example of this occurred with Amazon’s recruiting algorithm. The company realized after several years their program for selecting and hiring software developers favored men. This was because Amazon’s computer systems were trained with a dataset containing resumes from mainly men.
Because of this, their algorithm penalized resumes that included the word “women’s” as in women’s chess champion. And it downgraded an applicant if they had graduated from an all womens’ college. Amazon ultimately abandoned the program because even with editing, they could not make the program gender neutral.
Selection Bias Ignores the Real Population
In selection bias a dataset overrepresents one certain group and underrepresents another. It doesn’t represent the real population. For example, some machine learning datasets come from scrapping the internet for information. But major search engines and the data in their systems are developed in the West. As a result, algorithms are more likely to recognize a bride and groom in a western style wedding but not in an African wedding.
Can Big Tech Really Self -Police
Researchers are just beginning to understand the effects of bias in the machine learning algorithms. And the big tech companies which create these systems have pledged to address the problem. But others question their ability to self-police. Google recently fired an expert, vocal, high profile employee who they hired to focus on ethical AI. She was concerned about problems in the language models they used. This raises the point that ethical AI has to mean something to the most powerful companies in the world, for it to mean anything at all
The Power of Diversity
So, what can we do about algorithms which judge us and make decisions about us at every stage of our life, without us ever knowing? Experts say we need to be aware of the problem. We need to ensure the datasets are unbiased. We should develop and use programs that can test algorithms to check for bias. And a recent study emphasized that if the people training the systems come from diverse backgrounds, there is less bias.
We know data scientists inject their bias into the algorithms they build. Having diversity means the algorithms are built for all types of people. We’ve come to learn we need AI ethics because as one headline put it, “We Teach AI Systems Everything Including Our Bias.”
Thanks for listening, I hope you found this helpful. Be curious and if you like this episode, please leave a review and subscribe because then you’ll receive my podcasts weekly. From Short and Sweet AI, I’m Dr. Peper.
How fast can you develop a vaccine? Never has this challenge been put to the test quite so intensely as in 2020.
In fact, Jason Moore, who heads Bioinformatics at UPenn thinks that if the virus had hit 20 years ago, the world might have been doomed. It’s only thanks to modern technology that we now have a safe vaccine. He said, “I think we have a fighting chance today because of AI and machine learning.”
So, how did AI help to make the Covid-19 vaccine a reality? The short answer is a combination of computational analysis and the system of AlphaFold. I talk more about how researchers developed the vaccine so fast in this episode of Short and Sweet AI.
In this episode find out:
Important Links & Mentions
Resources:
Episode Transcript:
Friends tease me because I’m so fascinated with artificial intelligence that I will claim AI is the reason we have a safe Covid-19 vaccine so quickly. And they’re right, it is one of the reasons. In fact, Jason Moore, who heads Bioinformatics at U Penn thinks if this virus had hit 20 years ago, the world might have been doomed. He said “I think we have a fighting chance today because of AI and machine learning.
How did AI help to make the Covid-19 vaccine a reality? The short answer is through computational analysis and Alpha Fold.
But first, a little background on vaccines. A vaccine provokes the body into producing defensive white blood cells and antibodies by imitating the infection. In order to imitate an infection, you need to find a target on the virus. Once you find the target you need to understand its 3D shape to make the vaccine against it. But it’s really hard to figure out all the possible shapes before you find the one, unique 3D shape of the target, unless…unless of course you use AI.
In the case of the Covid-19 vaccine, Google’s machine learning neural network called Alpha Fold saved the day. Alpha Fold predicted the 3D shape of the virus spike protein based on its genetic sequence. And did it really fast, as early as March 2020, three months after the pandemic started. Without AI, it would have taken months and months to come up with what the best possible target protein could be, and it might have been wrong. But with AI, researchers were able to race ahead to ultimately develop the mRNA vaccine.
It’s common knowledge that it can takes years or even decades to develop a vaccine. Before Covid-19, using other approaches, the quickest vaccine to be developed took 4 years. As of September 2020, there were 34 different Covid-19 vaccines being tested in humans. That’s an astonishing number in so short a time.
Neural networks excel at analyzing massive amounts of data to find patterns that humans might not spot. Computers use machine learning to sort and analyze incredible amounts of data to learn and train over time. And that’s been AI’s second big contribution to conquering Covid-19. It’s called computational analysis. It involves using AI to gather insights from huge sources of experimental, and well as real world data, on the virus.
At the outset of the pandemic The Allen Institute for AI started an online repository of research articles about Covid-19. Today it has over 30,000 academic articles. Researchers can use this data set for the machine learning algorithms to train on, so they better understand the virus.
For example, as early as April 2020, computational scientists harnessed neural networks to sort through medical records by the thousands. The machines were able to confirm the lack of smell and taste is one of the earliest symptoms of Covid infection. There existed isolated reports of anosmia, which is the medical term for loss of smell and taste, but computer data analysis validated the finding. The CDC then added these to their list of Covid symptoms which helped identify when a person had the infection.
In another instance, medical charts from 96 hospitals in several different countries were analyzed with machine learning. What emerged was insight that many Covid patients had really off the chart readings of blood clotting. This alerted doctors to use blood thinners in patients hospitalized with Covid.
As scientists explain, the human brain becomes pretty quickly overwhelmed by the endless combinations of things, but when you use AI, the machines can find and directly move in on important findings, very quickly and effectively. AI is routinely depicted as evil in fiction, social media, and by Hollywood, and yet, its revolutionized how vaccines are created. It’s also become a workhorse of this pandemic as a powerful technology for processing massive amounts of information. Maybe, the machines will save us.
Technology breakthroughs are disrupting every industry at a rapid rate. In fact, advances in technology are massively transforming every industry exponentially faster than ever before in history.
What do you call exponentially fast disruption and massive transformation in worldwide industries?
It’s called the 4th Industrial Revolution, which I talk about in more detail in this episode of Short and Sweet AI.
In this episode find out:
Important Links & Mentions
Resources:
Episode Transcript:
Welcome to those who are curious about AI. From Short and Sweet AI, I’m Dr. Peper.
Right here, right now, technology breakthroughs are disrupting every industry and massively transforming every industry, exponentially faster than ever before in history. What do you call exponentially fast disruption and massive transformation in world-wide industries? It’s called the 4th industrial revolution.
The 4th industrial revolution is also known as 4 IR or Industry 4.0. But what does it mean? Klaus Schwab, founder of the World Economic Forum, coined the term and wrote a book of the same title. He details how we are now living during a 4th industrial revolution characterized by the fusion of AI, robotics, 3D printing, IOT, quantum computing, blockchain, autonomous vehicles, 5G, synthetic biology, virtual reality, and countless other technologies. He describes this as a “technological revolution… that is blurring the lines between the physical, digital and biological spheres”.
Technology merges with humans as our smart watches monitor our hear rate, our temperature or how much we move. It embeds in our daily lives as facial recognition, voice activated assistants, or apps on our phone. This isn’t the future, this is happening now. It’s changing how we live and changing who we are.
The three previous industrial revolutions also had new technology which fundamentally changed society. And yet, they were different. Let’s go back and look.
The First Industrial Revolution occured in 1760 with the invention of the steam engine and led to factory manufacturing. Hand-made goods were replaced by mass produced products. And the agricultural society was replaced by a huge migration to the cities. The Second Industrial Revolution came in the late 1800s with inventions such as the internal combustion engine, the lightbulb, the telephone and major infrastructure such as railroads as well as the steel, oil and electricity industries. The Third Industrial Revolution began in the 1960s with the invention of the semiconductor, personal computers and ultimately, the internet.
Schwab rejects the idea these present-day developments are part of the third industrial revolution. Four IR is evolving superfast, at an exponential, not linear pace, like the previous IRs. For example, it took 75 years for 100 million people to have a traditional telephone, but it only took 2 years for 100 million people to sign up for Instagram and less than a month for 100 million people to use Pokémon Go. The 4th industrial revolution involves many, many different technologies. Those technologies are combining and merging together and can transform entire systems, across companies and industries, and across cultures and countries.
What are the pro and cons of this new Industry 4.0? Advocates point out the increased productivity from technology and the improved quality of daily life where we can have almost anything we want on demand. There will be massive new markets created as more people come online. And more entrepreneurship exploding worldwide as barriers to new businesses are lowered.
But many thoughtful people are concerned about the cybersecurity risks as everything becomes so connected through the IOT. And disruption of core industries has already begun with Airbnb challenging hotels, Uber and Lyft dissolving the taxi industry, and Amazon threating any business that sells, well, anything. There are ethical concerns about access to data on individuals or groups being wide-spread and used for personal gain and manipulation.
But perhaps the greatest threat of the 4th industrial revolution is the specter of massive inequality. Experts fear there will be a divide of high-skill/high-pay workers and low-skill/low-pay workers in a winner-take-all economy, as the middle-class dissolves. Even Schwab predicts that inequality will be the greatest concern affecting society in the 4th Industrial Revolution.
Typically, early adopters of new technology gain the greatest financial benefits, allowing them to jump ahead, while the income gap widens. Sounds pretty dire and yet, no one knows. The French philosopher Voltaire said, “Doubt is an uncomfortable condition, but certainty is a ridiculous one.” This revolution is creating change at warp speed. And even those with knowledge and preparation may not be able to keep up with the ripple effects from the changes.
Is it time we regained control of our data and found new and better ways to protect it?
You and I know that the social media platforms and internet sites we visit collect data on us. In many ways, they monetize our data and use it as a product that can be purchased.
In this episode of Short and Sweet AI, I talk about personal data as private property and whether there is a way for us to choose who gets to use our data.
In this episode find out:
Important Links & Mentions
Resources:
Episode Transcript:
From Short and Sweet AI, I’m Dr. Peper, and today I want to talk with you about personal data as private property.
You and I know that social media platforms and internet sites we visit are collecting data on us. We know they’re selling our data to advertisers. I mean, that’s their business model. They provide a platform for us to connect with each other and we give them our personal data as payment. Data is valuable. Data is the new oil. It brings in billions of dollars of income for Google, Facebook, Instagram, Amazon, and countless other companies. When we’re online and we click on a pop-up that says “accept”, we’re essentially giving away our personal information to that company. And do we really have a choice? You either have to accept the terms or you’re not allowed to use that site.
Well, what if we could be paid for our data, what if we could determine who gets data about what sites we visit, what apps we use on our phones, what physical locations we go to, what conversations we have, basically what if we could be paid for all the information companies are gathering on us now on a daily basis. And what if we had a system that only provides our data to who we say with great privacy protection using the security of a block chain type technology. Enter Professor Dawn Song and her company Oasis and we are one step closer to that reality.
Professor Song is considered to be one of the world’s expert on computer security. She is a Mac Arthur “genius’ recipient and a professor at UC Berkley. Much of her work is in the area of machine learning which I’ve talked about in a previous podcast and in adversarial AI. Adversarial AI is the study of how computer systems are hacked to transmit the wrong information.
While still a graduate student at Berkeley, her research drew attention for showing machine learning algorithms can infer what someone is typing. She showed hackers could use software to figure out someone’s password from the timing of their keystrokes picked up by eavesdropping on a network. Professor Song and her students were also the first to demonstrate that computer vision can be fooled. She applied a few benign looking stickers to a stop sign. As a result a driverless vehicle identified the sign as a 40 mile per hour speed limit sign instead of recognizing it as a stop sign and continued through an intersection without stopping.
She began by showing that a lot of these machine learning algorithms have weaknesses and she became passionate about people having control over their personal data. Her expertise in machine learning, computer security, and blockchain gave birth to Oasis Labs. She describes Oasis as a privacy-first, cloud computing platform on blockchain. She is creating technology which empowers users to protect their personal information, to decide who can use it, and to get paid for their data.
Through a program with Stanford Medical School, patients can use the Oasis platform to decide who to share their medical data with and to get paid when it’s used. They agree to have scans of their retina and other medical data shared privately through a blockchain type application on the Oasis platform. And then researchers use this information to train computers to recognize eye diseases. Meanwhile Nebula, a genomics company, is jumping onboard and has integrated with Oasis to give users control of their personal genomic data.
Professors Song’s vision for the future is for people to have a revenue stream from their personal information. It may not be a lot on a monthly basis but could contribute to retirement savings as companies pay for using your data over your lifetime. As she says “Today, companies are taking users’ data and essentially using it as a product: they monetize it. The world can be very different if this is turned around and users maintain control of the data and get revenue from it.”
This is a really revolutionary idea. Professor Song has created an internet platform which uses blockchain technology to give us the ability to control our data and earn an income from it. Personal data as private property, I think it’s time.
If you like these flash talks, please leave a review and subscribe. From Short and Sweet AI, I’m Dr. Peper
In this exciting episode of Short and Sweet AI, I talk about the recent update that Elon Musk gave on his company Neuralink – including how and why his team implanted a coin-sized computer chip in a pig’s brain to create a brain-to-machine interface.
In this episode find out:
Important Links & Mentions:
Resources:
What does it take to be the godfather of AI? And, how does someone come to obtain such a legendary title?
In this episode of Short and Sweet AI, I talk about Geoffrey Hinton, a neuroscientist, computer scientist, and the man Google hired to make AI a reality. In many ways, we have Geoffrey Hinton to thank for developing modern AI and deep learning. It is thanks to him that deep learning has become mainstream in the field of artificial intelligence.
So, how did Geoffrey Hinton rise to become the godfather of AI? Watch this video to find out!
In this episode find out:
Important Links & Mentions
Resources:
Moore's Law is coming to an end, and many people don't know how to feel about it. In fairness, the end of Moore's Law is not something that crept up on us out of nowhere. Industry experts predicted the termination of Moore's Law years ago. They observed its gradual decline and forecasted a grim future for Moore's Law that has since proved to be an accurate calculation.
But the question remains… why is Moore's Law ending? And why should you care?
I'm kicking off the start of the year with a Short and Sweet AI podcast episode that focuses on endings. That is, the end of Moore's Law and why it matters. As always, I focus on AI in simple terms so that whether you're new to AI or a seasoned pro, you can follow along fully immersed!
In this episode find out:
Important Links & Mentions
Resources:
The world's most powerful supercomputer would take 10,000 years to solve a math problem a quantum computer solved in minutes. Welcome to quantum computing.
The post What is Quantum Computing? part 2 appeared first on Dr Peper MD.
Quantum computing is an extraordinary technology based on quantum physics which uses quantum bits or qubits to solve problems in a magical way.
The post What is Quantum Computing? appeared first on Dr Peper MD.
I've interrupted my podcasts to care for patients during the COVID surge in my area. Many death certificates, many flags at half-mast.
The post A Physician during COVID appeared first on Dr Peper MD.
Our TOPPODCAST Picks 
