Dualities of AI
We have this incredible opportunity with AI, because of the way it’s trained to look at ourselves and our society. AI is a mirror, not a master. Tim O’Reilly
Bright side
We are not pursuing AI to beat humans at games. We are pursuing AI so that we can empower every person and every institution that people build with tools of AI, so that they can go on to solve the most pressing problems of our society and our economy. Bill Gates (The Economic Times 2016)
Exercise — Level-up
Read following article “Everyone is above average” (Mollick 2023).
Take 10 minutes to read the article.
Afterwards we will discuss following questions:
- Do you think that AI is rather an escalator or a king-maker? Why?
- What are risks of AI taking that role?
- What could be broader, long-term socioeconomic implications of AI leveling skills?
- What other effects might AI have on a societal level?
- What can be done to maximize the benefits and minimize the risks?
Dark side
It seems probable that once the machine thinking method had started, it would not take long to outstrip our feeble powers … They would be able to converse with each other to sharpen their wits. At some stage therefore, we should have to expect the machines to take control. Alan Turing
Case Study
Form groups of three, take 20 minutes to revisit the case study and work on following questions.
- Do you think that we will see something like artificial general intelligence in our lifetime?
- Which of the aforementioned risks worries you the most? Why?
- What is your opinion on the design of the OpenAI organization regarding its ability to deal responsibly with AI?
- Do you think the self-regulation approaches taken by organizations like OpenAI, Microsoft, Meta, and Google are sufficient? Why (not)?
- What do you think about making LLM and other AI technologies available via open source?
- What are benefits and risks of legal regulations regarding the design and use of AI?
Responsible AI governance
I’m increasingly inclined to think that there should be some regulatory oversight, maybe at the national and international level, just to make sure that we don’t do something very foolish. I mean with artificial intelligence we’re summoning the demon. Elon Musk
Let’s do some research to learn more about private sector-driven approaches to moving toward responsible AI.
Exercise — Google and Microsoft
Research about either
Take 20 minutes to understand the standards/principles, evaluate them critically and create a short summary of your learnings (approx. 3 minutes presentation)
Exercises
EU AI Act
Research about the Artificial Intelligence Act of the European Union.
Try to understand the risk-based approach and provide a short summary, also tackling following questions:.
- What are high risk AI systems?
- Which requirements should be mandatory requirements for these?
Exercise — Fairspeech
Have a look at the fairspeech website (and at the research paper (Koenecke et al. 2020)).
Tackle following questions:
- What is the issue with current automated speech recognition (ASR) systems ?
- Which ethical issues do they cause?
- What should be done to mitigate the risks?