There's been a lot of talk about the best way to apply ethics principles to the AI and machine learning algorithms that influence our daily lives more and more. With a growing number of frameworks developed around this question, it may be intimidating for a founder or technical professional to even start approaching this topic in their own work.
In an attempt to solve this problem in a hands-on way, the SHOP4CF project has launched a massive open online course (MOOC) entitled Human-Centered AI & Ethics, available for free to anyone interested.
As the description of the course states, “AI ethics is a universal tool that helps developers, innovators, scientists, researchers and leaders to better connect technology development, policies and human rights, algorithms and societal criteria / logic, data and social sciences. The key objective of this course is to provide you with such a universal tool and ecosystem — human-centred AI-framework, which you can build, implement and keep updated for your product or organization, that meet requirements of responsible governance, transparency, algorithmic diversity and fairness.”
Acknowledging a problem is the first step to solving it
In case you've missed our previous coverage of SHOP4CF (stands for Smart Human Oriented Platform for Connected Factories), it's an EU-funded project that aims to create a unique infrastructure for the convenient deployment of human-centric industrial applications. One of its goals is to create a “GitHub for manufacturers” in order to make smart factory technology accessible for SME-level players.
The project is now working with its first batch of tandems of a manufacturer and integrator, who can get €100,000 in grant money to introduce automation on the production floor using components developed by a number of industrial partners.
At the same time, SHOP4CF strives to raise awareness of the human-centred AI approach and give the broad community the right tools to apply these principles in practice through the freshly-launched MOOC.
“ we talk about implementation of social robots, cameras, staff education and VR training, and so on — and all this systems involve AI in some way,” said the AI ethics and human rights expert Yonah Welker, who authored the course. “So, we encompass the entirety of the problem and explore how to make these systems better, using existing cases.”
Thinking beyond manufacturing and IoT
The course, however, is designed to be interesting and useful for anyone whose work is connected to AI and machine learning in any way.
“On the one hand, I introduce the AI ethics principles, but also the whole ecosystems of organisational frameworks we have today,” Welker said. “I also try to reflect the key challenges we have today in 2021, like fake news, lack of genuine data, irrelevant training data, problem of accountability, involvement of autonomous agents, and so on. After that, I suggest creating a framework of how we can solve it right here and right now.
“So, the third part of the course is actually the creation of some kind of a scheme of a framework, step-by-step. The students can work on it in their organisations in order to fix this problem, to create the culture ecosystem vocabulary, build transparent AI, build end-to-end accountability for agents, developers, and designers involved in the process.”
Blazing the AI ethics trail
An important part of the course is a look at the regulations across the world applicable to tech in general and AI in particular.
“In most cases, the problem I see is two extremities,” Welker explained. “Some companies would love to use everything without any kind of restriction like facial recognition, computer vision, etc.; another extremity is the idea that we need to limit everything. So, in this course I share existing cases and ways how we can do it in a smart, flexible way for educational solutions, classroom workplaces, specifically for health and future of learning.”
In Welker's opinion, Europe is uniquely positioned to become the world leader in AI ethics research and regulation thanks to the interest in this area from both practitioners and regulators.
“In the US, most of the AI ethics professionals are in academia or in private companies,” he said. “In Europe, we also have policy makers who care about it. Some pieces of legislation are not ideal, but at least these policy makers have proposed them. In addition to that, we have a community and a discussion at the same time. The government itself is more on the society side and cares about protecting the people.”
To deepen your knowledge of AI ethics and stay in the know about the recent developments in the field, enrol for SHOP4CF's free course today at shop4cf.mooc.ramp.eu.