AppliedAI

Despite the rapid evolution of AI, projects still fail at a disappointingly high rate. In the past, capturing data at scale and building models was the challenge, but today we're confronted with the issue of making AI more robust while avoiding the risk of unintended consequences. While the tools are new, many challenges remain the same. In this talk, I will share real-world stories and applied examples that demonstrate:

• How to build the business case for an AI project (and get buy-in)
• Navigating AI project management to prevent failure
• How to mitigate the risks of unintended consequences from using AI

Our presenter this month is Cal Al-Dhubaib. Cal is a data scientist and entrepreneur, with an emphasis on designing and building trustworthy AI. He empowers organizations to launch data science initiatives that grow the bottom line. He is especially passionate about the ethics of AI and how organizations can build the right talent to support AI initiatives.

Cal has received awards at both the national and international levels for his work in predictive modeling and entrepreneurship and has been listed on Crain's Cleveland 20 in their 20s and Cleveland Smart 50.

In the midst of implementing the latest AI technology, it can be easy to miss how lay people react to this. Spoiler alert: Confusion and anxiety. How do you curate their experience while you're heading full-tilt toward this brave new AI future?

Join our speaker Peter Scott, as we explore some of the many things technologies need to consider as they implement Artificial Intelligence.

Peter is the author of "Crisis of Control: How Artificial SuperIntelligences May Destroy or Save the Human Race" (Nitrosyncretic Press, 2017) about the existential risk that humanity faces from exponentially advancing technology. He is a certified coach (International Coaching Federation) and focuses on coaching organizations and individuals to prepare for radical change.

He also has many years of IT expertise in enterprise architecture, business solution consulting, system engineering, and systems analysis, and has written several instructional books and videos.

You can find more about Peter at: https://humancusp.com/

NOTE: Meetup is IN-PERSON at Lab651 in Saint Paul, MN. We look forward to seeing you at the meeting. Please register so we can plan for food and beverages.

Members of Modal Technology Corp and RISC AI, Inc., both Minneapolis, MN-based companies, will present the status of commercializing ALiX - a disruptive AI training technology that is non-statistical and guarantees the optimal solution (global min) in a single run. A by-product of the training is output which is explainable without any additional work required. ALiX promises to reduce the complexity of models, the compute resources needed, and organizational staff requirements for training AI models. The benefits of reducing model complexity translate to fewer compute resources for inferencing once models are deployed. The technology is proven in SW emulation mode; a real-world case study will be presented related to the identification of key biomarkers for colo-rectal liver cancer, in a project conducted between Modal Technology and McGill University Health Center. A RISC-V processor is currently under development by RISC AI to bring enterprise-scale training with ALiX to the market by Q[masked]

Containers are a great way to build applications, but they’re not always easy to make or use—especially for data scientists. Chassis is an open-source project that turns ML models into containerized prediction APIs in just minutes. We built this tool for Data Scientists, Machine Learning Engineers, and DevOps teams who need an easier way to automatically build ML model containers and ship them to production.

Presenter:
Brad is an experienced technologist with a focus on AI/ML & Machine Learning Operations (MLOps). As a Data Scientist and Solution Engineer at Modzy, I help organizations deploy ML models at enterprise scale and unlock value in their AI investment projects through the implementation of robust MLOps pipelines.