Youtube link: https://youtu.be/Tb3KOlQIlgA
Slides:
Register in advance for this meeting: <deleted as event has passed>
After registering, you will receive a confirmation email containing information about joining the meeting.
Title: Artificial Intelligence is on a Rampage: A Call for Guardrails & Regulations
Speaker: Peter MacKinnon, Senior Research Associate, Faculty of Engineering, uOttawa & Member IEEE-USA Artificial Intelligence Policy Committee, Washington, D.C.
Abstract: The AI rampage started in the public eye less than a year ago with the release of an AI tool based on advances in natural language process (NLP), a branch of AI, and generally referred to as a chatbot. ChatGPT, the tool in question, is now in its 4th iteration since release in November 2022, with each iteration improving performance and its anthropomorphic likeness. ChatGPT currently has over 100 million monthly users and the website generated 1.6 billion visits in June of this year.*
Concurrently, other types of AI tools that manipulate and replace a person’s likeness with that of another have created a burgeoning field of deepfakes. This is achieved through the use of generative visual and audio content by means of AI methods based on deep learning and neural network architectures, such as generative adversarial networks. We will highlight these and other technical developments.
Current uses of AI in the form of chatbots, deepfake imagery and creation and manipulation of sound (e.g., a person’s voice), are creating a wide range of concerns about the design, deployment and use of such sophisticated AI tools. Consequently, governments, lawmakers and many commentators on the evolution of AI are raising concerns these days about the need for AI guardrails and regulations.
We will explore the present state of national and international efforts to constrain unrestricted competition through various ways and means that amount to at a minimum guardrails, (e.g., voluntary buy-in to certain constraints like 3rd party certification of algorithms) to hard rules with criminal penalties. * https://explodingtopics.com/blog/chatgpt-users
Biography: Peter has a background as a scientist, business manager, entrepreneur, domestic and international bureaucrat, executive, diplomat, management advisor, and academic; including today being affiliation with both the Telfer School of Management and the Faculty of Engineering at the University of Ottawa.
He is a pioneer in the commercialization of AI and today is actively involved in ethical and policy issues related to AI, especially as a member of the IEEE-USA Artificial Intelligence Policy Committee. Peter has an extensive background on the forefront of scientific and technological breakthroughs around disruptive technologies and their impacts on society. See www.linkedin.com/in/peter-mackinnon-ba88682a/ for more background.
Peter, as noted, is associated with the Faculty of Engineering at uOttawa and serves as the Chair of the Foresight Synergy Network, sponsored by the Telfer School of Management at uOttawa.
Recently, Peter penned an op-ed with Eli Fathi, Chairman of MindBridge Analytics, an AI firm involved in financial fraud detection and recipient in 2022 of the Order of Canada. The topic of the op-ed is Generative Artificial Intelligence & Integrity: Winners and Losers. Access to this piece will be provided during the webinar.
Foresight Synergy Network (FSN) 