Far from the popular, and often fictional, ideal; Artificial Intelligence is part of people's lives. Today we speak with an expert in this technological field: Mónica Villas is an industrial engineer, specialising in AI, Cloud, Blockchain, Analytics, and IT.
An extensive CV, which also highlights her more than 20 years as a manager at IBM, as well as her experience as a lecturer at various institutions and her role as a collaborator with several organisations.
For over two years, Mónica has been part of the teaching staff at IMMUNE. She is our Artificial Intelligence Knowledge Leader.
To begin with, do you think society is aware of all the AI surrounding it?
I don't think so. None of us. In fact, in our courses, it's one of the first questions we ask: “Do you use Artificial Intelligence?”. And it's very surprising when half the class, or even almost no one, says they don't use Artificial Intelligence.
But, from the moment we get up in the morning, we have Smart alarm clocks which know how to wake us up at the time of day when we’ve had enough sleep (it’s no longer at a specific time).
Then, if we continue, we get up and look The weather. That time is a prediction: we know what it does today, but we don't know what it will do in 3 days.
Or if we use any of our apps for booking a taxi or transport, they do a predictive analysis and assign us that vehicle, based on how far away it is. Or if any of you have a Rumba, one of these devices that automatically vacuum the house, also has its AI algorithms where they predict which parts of the house get dirtiest, etc.
Every day, we are surrounded by AI or algorithms that use AI. It's not the future, it's the present.
2. How has AI evolved in recent times?
AI is from the 1950s. From the important milestones in that era, we have moved on to having that famous chess match between Kasparov, which the IBM machine, Deep Blue, won in '97; but, Really, when AI has evolved is in the last 10 years.
It has evolved a lot due to the data paradigm we have now. Infrastructures have changed, the way of computing has changed.
We have the Cloud available, we have infrastructures that are already capable of performing very advanced calculations and not at a very high cost.
And it has also been thanks to Now we have algorithms available to everyone in the world, which everyone can use.. In the last ten years, we've seen the autonomous car make use of AI capabilities, as another new AI machine, Google DeepMind, beat the best Go player, and we've seen the first painting made by AI. The advancements are very important.
3. A European Digital Strategy, the White Paper on Artificial Intelligence, the European AI Regulation… these are just a few examples with which the European Union has set some guidelines for AI. Do you believe that Artificial Intelligence must or should have limits?
Without a doubt, Artificial Intelligence must be ethical. And I am not alone. As we have seen, Europe is leading this change towards ethical Artificial Intelligence. Responsible AI.
It's true that it's a technology that will help us to automate and improve our businesses. But, on the other hand, it's a technology that makes automatic decisions. That is the key point that has prompted the EU's reaction and this regulation, which has been a proposal since April 2021 (and has been in development for 3 years).
It is a privilege for those of us working in Europe that this Artificial Intelligence regulation thinks about ethical and responsible AI.
We need to create ethical AI that is free from bias. Well-known things like Amazon's algorithm that only selected men for hiring, or certain decisions being made based on errors that could have been foreseen, cannot happen.
4. In your opinion, what awaits us from AI in the future?
I would like to have that magic ball to know what awaits us in the future. I believe many changes are being sighted, particularly in the area of natural language. AI is made up of Machine Learning and Deep Learning. And Deep Learning is the area that is growing the most.
The natural language area with GPT3 – and we're already on GPT4 – is a Deep Learning algorithm with billions of parameters that is capable of recognising natural language with ever-increasing accuracy.
This is going to help us improve our Spanish, for example. Maria From MareNostrum Artificial Intelligence, which is led by SEDIA, it also helps the Spanish language to be better understood by machines... I think they will go in that direction.
Also, we have to take into account the quantum computer. We are looking at the future with the eyes of computers as we know them now. Quantum computing will, in some way, bring us better computing times and allow us to solve problems that we haven't been able to solve with traditional computing up until now.
Another area where AI is being used a lot is in Medicine. Last year, Google's DeepMind, in the Project AlphaFold, It predicted how proteins are able to come together, how amino acids join to form a protein. This - in medical terms - used to cost thousands of euros and years. And we've moved to shortening this process.
5. Data Science is, without a doubt, another of the booming tech sectors. What kind of professional profile can enter the so-called “data science”?
I don't think there's a typical professional profile. The DATA SCIENCE It is made up of 3 interconnected circles. Technology, on the one hand, referring to all Machine Learning algorithms.
Then we have a part which is the Mathematics and Statistics, which is important. And the last one – which we often forget – which is about the industry. The marketing industry isn't the same as retail or the electricity sector… And AI can help in all.
These 3 parts need to be combined to get a data scientist.
Who is the profile? We have marketing profiles who come to learn more about the technology side, just as we have technology profiles who want to learn more about algorithms or computing.
Data science might indeed be better positioned for those coming from the tech world, but right now it's open to anyone who wants to leverage data to improve their business.
And what kind of companies demand data scientists?
Any company that wants to unlock the potential of its data to improve its business. The Data Never Sleeps report, which states we generate 2.5 trillion data points daily, highlights the need for this technology. I can't extract data manually. I need to learn how to harness this Big Data and apply AI to it.
Where is this valid? In marketing, to improve online marketing or its automation and see which products to recommend, or also in the detection of anomalies in an X-ray, or in the predictive maintenance of machines, or for the prediction of electricity demand... In short, Any company that has data.
Furthermore, a recent report – which discusses the jobs of the future – places the data scientist in second position, after the need for professional profiles to support online sales.
For example, we could have a data scientist who knows how to interpret all that online and offline data that we receive from our customers to make the best possible offers or to make the most of that data we have and personalise the offer.
7. Given your extensive experience in the sector, what would you tell a future student who wants to train in Data Science or Artificial Intelligence?
It's not just from my experience, if you read the LinkedIn, InfoJobs, and Adecco Business Insider Report, which talks about the jobs of the future, those jobs include: The Big Data specialist, the Cloud specialist and the Data Scientist.
It is these tech profiles who will be able to access this data. Some will be more technical to analyse in a bit more detail (Big Data), others to know where to put that data (the subject of Cloud), and the data scientist will want to extract value from the data.
What I recommend to profiles is that they never lose that business side. Data science roles are those that seek to solve a business problem using the data I have. That data will be in the cloud and will be protected for reasons of cybersecurity or perhaps Big Data; but let's never forget the business side.
This is why, in our training... IMMUNE, we combine the technological part (all that knowledge necessary for the data cycle) with a business part, which are our “Industry Talk” sessions (knowing what's happening in the market); where experiences in the industry with AI are discussed.
