Lernkarten

Karten 88 Karten
Lernende 2 Lernende
Sprache English
Stufe Universität
Erstellt / Aktualisiert 15.01.2020 / 21.01.2020
Lizenzierung Keine Angabe
Weblink
Einbinden
0 Exakte Antworten 87 Text Antworten 1 Multiple Choice Antworten
Fenster schliessen

What does AI actually mean?

Artificial intelligence (AI) refers to systems that display "intelligent" behavior by analyzing and taking actions - with some degree of autonomy - to achieve specific goals

 

- intelligent behaviour

- analizing

- taking actions

- degree of autonomy

Fenster schliessen

What are the UN Sustainable Development Goals?

UN Suistainable Development Goals are the bluepring to achieve a better an more sustainable future for all. They address global challenges we face, including those related to poverty, inequiality, climate, environmental gedradation, prosperity (Wohlstand) and peace and justice. 
 

The goals interconnect and in order to leave no one behind, it is important that we achieve each goal and target by 2030.

 

Examples

- Donate what you do not use (no poverty)
- waste less food (zero hunger
- vaccinate (Health and wellbeing)
- Educate childern
- Gender equality, empower women and girls
- Avoid wasting water
- affordable and cleand energy (use only energy-efficient bulbs)
- create job oppurtunities for youth
- fund projects that provide basic infrastructure
- reduce inequalities
...

Fenster schliessen

Differences in AI in China, US and Europe?

USA: most of the AI companies are based in de US. Start-ups, talent attractor, private equity (Gerechtigkeit)

Europe: loses talents to the US but has high quality AI researchers, but struggles to translate research into business application

China: rapidly catches up, leads in adoption speed and available data

Fenster schliessen

What is the European way to develop AI?

- The EU coordinated plan on AI states that its "ambition is for Europe to become the world-leading region for developing and deploying cutting-edge, ethical and secure AI

- Positioning AI made in Europe as a positive global brand

- combining innovation-freindliness and ethics: AI as economic value

Fenster schliessen

What is written in the EU Ethics Guidelines for Truthworthy AI?

The guidelines were created by the high-level expert group on AI (52 pax), representing a wide range of stakeholders (research, industry, NGOs, ...).

AI is not an end in itself, but a tool that has to serve people with the ultimate aim of increasing human well-being.

Trust is an important prerequisite to ensure a human-centric approach to AI.

The European Union is founded on the values of respect for human dignity, freedom, democracy, equality, the rule of law and respect for human rights, including the rights of persons belonging to minorities: AI applications must be consistent with these values.

Everyone has the right to own AI or share them. Nobody should take our things from us without a good reason

Copyright is a special law that protects AI. Others cannot make copies without permission. We all have the right to our own way of life and to enjoy the good things that art, science and learning bring.

Fenster schliessen

7 key requirements for ethical AI?

  • Human agency and oversight
  • Technically robustness & safe
  • Privacy and data governance
  • Transparency
  • Diversity, non-discrimination and fairness
  • Societal and environmental wellbeing
  • Accoutable
Fenster schliessen

7 key requirements for ethical AI.
What means Human agency and oversight?

Key aspects

  • Fundamental rights (liberty, freedom of thought, freedom of expression, freedom of movement). Mechanisms should be put into place to reveive feedback regarding AI systems that potentially infringe fundamental rights
  • Human agency (Users should be able to make informed autonomous decisions regarding AI systems. They should be given the knowledge an tools to comprehend and interact with AI systems to a satisfactory degree and where possible, be enabled toe reasnoably self-assess or challenge the system
  • Human oversight - may be achieved through governance machanisms such as human-in-the-loop (HITL), hum-onthe-loop (HOTL) or human-in-comman (HIC) approach
Fenster schliessen

7 Key Requirements for Trustworthy AI

Technical robustness & safety. What does it mean?

  • Resilience to attack security (AI like all softwares should be protected agains vulnerabilities, be exploited by adversaries, eg. hacking)
  • Fallback plan and general safety (AI systems should have safeguards that enable a fallback plan in case of problems. That can mean asking for a human operator before continuing their action
  • Accuracy (Accuracy pertains to an AI systems ability to make correct judgements for example to correctly classyfy information into the proper categories or its ability to make correct predictions