Worldwide AI Webinar 2022 brought together AI/ML executives and experts to share insights, best practices, and research on several thought-provoking and interesting topics. This conference took place over two days with 20+ speakers and thousands of attendees from all around the world.
This article will provide highlights of a few notable keynote talks and two insightful panel discussions.
The recordings will be shared shortly so please follow our social media channels to stay up-to-date.
How to create an AI mindset/AI literacy in a business
SAP’s VP of AI, Andreas Welsch, laid the foundation for day 1 by covering the essential points to create an AI mindset or AI literacy in business.
Sharing that about 54% of AI projects make it from pilot to production, Andreas Welsch pointed out that the key to AI adoption success in enterprises is AI literacy. He then highlighted the actions needed for executives to build an AI mindset within the organization, emphasizing choosing a product mindset over a project mindset.
Explainability and Ethics in Healthcare AI
This session by Patrick Bangert, VP of AI at SAMSUNG SDS, delivered was mind-boggling. He started off by stating that while a cancer diagnosis can take up to 6 weeks if carried out by a human doctor, an AI can do it instantly by turning images into sentences.
However, while machines are known to make fewer mistakes than humans, they can’t completely avoid errors. This is where, according to Patrick, data governance comes in. Should we want to integrate AI deeply into healthcare, regulators must think about its impact and work with researchers to come up with a well-rounded set of rules based on ethics frameworks.
Increasing Organizational Maturity by Adopting AI/ML through the End-to-end Lifecycle for Increasing Business Value
Ali Arsanjani, currently the director of Cloud Partner Engineering at Google Cloud, delivered the most insightful keynote. He began with the timeline of how more than 3000 Deepmind/Google researchers have built state-of-the-art AI that resulted in better product performance and lower costs for customers. He showcased the AI/ML digital innovation spectrum, explaining that organizational maturity will evolve with the ability to traverse through it.
Dr. Arsanjani then followed with the hidden technical debts in machine learning systems and 6 levels of ML adoption. Finally, he concluded his speech with 5 key components of the end-to-end ML journey. This is an excellent talk for anyone interested in learning more about end-to-end MLOps and managing ML models to increase business maturity and values.
Panel discussion: Challenges and Solutions for Adopting AI & ML in your Enterprise
The first panel discussion of Worldwide AI Webinar 2022 was an informative and hilarious one. Joining this panel was Ali Arsanjani, director of Cloud Partner Engineering at Google Cloud, Andreas Welsch, VP of AI at SAP, Aamar Hussain, Director of Azure Data at Microsoft, and Nestor Camilo, CASP Director at Oracle.
The session started with each of the speakers sharing their thoughts on some tough challenges enterprises face when adopting AI/ML and what the common processes to review, assess, and roll out AI/ML solutions for an enterprise are.
To lighten up the discussion, or as Dr. Ali Arsanjani put it “to entertain the audience at our expense”, the panelists challenged each other’s viewpoints to shed light on the current AI/ML adoption.
Then, all four speakers provided insights into an adoption challenge they thought was the easiest to overcome and how to overcome it. The panel discussion ended with each panelist sharing the best tip on how to drive the adoption of cutting-edge tech at a resistant enterprise.
Why data is at the core of AI/ML
Aamar Hussain, the Director of Azure Data at Microsoft delivered a wonderful speech on the vitality of data to AI/ML. He explained why we should maintain focus on data to build a more systematic and efficient system by comparing data as the main ingredient to ‘cook up’ an amazing AI ‘meal’. Aamar then shared his best practices and recommendations before rounding out the most effective data strategy for AI/ML.
How to build a responsible applied AI at scale
Noelle Silver, an award-winning technologist, and the AI Executive at IBM delivered a great speech on building responsible applied AI at scale. She covered enterprise AI scenarios and highlighted the importance of conversational AI, knowledge mining, and process automation. Noelle spoke often about her belief in open source and encouraged the audience to take a go at building their own no-code apps to make scaling responsible applied AI a reality.
NTIS – A Pathway for Data Innovation and AI in the U.S. Federal Government
Dr. Ola Olude-Afolabi, a data scientist at the U.S. Department of Commerce had an interesting chat with our audience about the use of data and AI in the US federal government.
She shared that having operated since 1988, NTIS has evolved over the years and morphed into creating innovations and AI solutions. With the new Joint Venture Partner (JVP) program, NTIS aims to offer the US government and its citizens solutions to pressing matters such as cybersecurity, fraud & anomaly detection, and enterprise data governance.
Dr. Olude-Afolabi specifically emphasized the importance of responsible AI, which is a core principle of the JVP program. While AI and data have been at the forefront, protecting citizens remains the nation’s top priority. Therefore, DOC encourages other corporate entities to keep financial data and confidential information safe and secure by joining the program.
Open Debate: Artificial General Intelligence
The session concluded Worldwide AI Webinar 2022 was a long-awaited one!
An open debate on Artificial General Intelligence with the participation of VP of AI at Samsung SDS, former Head of AI at Volkswagen, professor of AI at Oxford Brookes University, AI Executive at IBM, and Director of Engineering at Thomson Reuters promised great insights into the real picture of AGI and it delivered!
The panelists began the session by dissecting the real meaning of Artificial General Intelligence, whether it has to gain consciousness or develop its own purpose to be true AGI. They then discussed the obstacles that are preventing AGI from becoming imminent and what kind of test is needed for AGI. The four speakers proceeded to share their insights into whether AGI can solve any commercial applications.
The session ended with the whole panel answering a provocative question from the audience:
"What can AI comprehend that human beings can’t?"
Patrick Bangert pointed out that while machines can be more consistent and accurate than most human beings, it doesn’t mean they can comprehend a task better than we do. Fabio Cuzzolin shared that he was developing a theory around this topic and questioned the true meaning of consciousness in AGI. David von Dollen agreed and added that while the concern about AGI is real, he’d like the audience to think about the effects of weak AI on humankind.
What an exciting two-day event! If you’ve arrived at this conclusion, you have now understood better AI and its applications in multiple fields. Ready to implement the best practices outlined during this event within your organization? Contact Wow AI, a global provider of high-quality AI training data.
To view these talks, as well as the many others not mentioned in this article, follow Wow AI on LinkedIn
and visit our website where recordings will be available soon.
We hope you learned and enjoyed the event as much as we did.
See you at the next event!