General Insights - AI, transparency & the legal implications

The AI Sustainability Center, Chalmers AI Research Centre, and AI Innovation of Sweden welcome you to join the discussion on the challenges and opportunity with the upcoming EU regulatory framework, how organizations can be ahead of the regulatory curve, and dilemmas of AI transparency.

The EU recently released ethical guidelines for AI, and is currently elaborating on recommendations for future related policies with regards to legal issues related to AI.  Certain states in the US are also moving ahead with regulation designed to protect privacy. The regulatory landscape is continuing to evolve, and organizations need to follow developments carefully in order to be prepared.  

Transparency is often seen as a fundamental cornerstone of understanding data-driven technologies and a vital component of achieving accountability. The absence of transparency in connection with algorithm-driven processes, sometimes referred to as “black-boxing”, is a well-known problem, and the lack of transparency is often described in terms of a trust deficiency. On the other hand, too much transparency could also be a problem in certain cases.  It can be a fine line in finding the right balance of transparency to satisfy stakeholders but not revealing too detailed or confidential information.

As AI becomes more important in applications such as healthcare and transportation, the reliability, openness, and fairness of these systems are increasingly important. As a company, trust is the new currency. In conjunction with society’s increasing use of, and dependence on, AI and data-driven technology, there is a growing need to understand potentially negative consequences and risks. The most advanced forms of AI, such as deep learning, may be the most valuable from an economic and societal perspective, but is also the most difficult for humans to understand. As AI and data-driven technologies become more advanced, it will result in trade-offs between transparency and accuracy.

Legal and ethical frameworks, standards, certifications or procedural stances will be necessary in order for companies to ensure AI is developed and implemented in an ethical way, as well as to maintain trust in both the technology and the organizations utilizing it. The seminar aims to shed a light on recent developments.

To programme and registration
Category Event
Location: Lindholmen Science Park, Lindholmspiren 5, Gothenburg
Starts: 03 December, 2019, 14:00
Ends: 03 December, 2019, 16:30

Published: Mon 11 Nov 2019.