AI continues to transform our world as companies look to win over consumers with intelligent experiences delivered in real time on smartphones, smart TVs, smart cars—smart everything.
Copyright: forbes.com – “Top Six Trends (And Recommendations) For AI And ML In 2023”
But along with new opportunities, organizations are also finding new challenges as they seek to cross the AI chasm. Here are the top six AI/ML trends that I’ll be tracking in the year ahead, along with recommendations for how enterprises can stay ahead of each trend.
1. Real-time use cases drive changes in the ML tech stack
A recent study by our company’s research group, Verta Insights, found that more than two-thirds of ML practitioners expect real-time use cases to increase significantly over the next three years. This trend will challenge companies whose ML tech stacks were built around analytical/batch workloads that are not suited for operationalizing real-time use cases at scale in customer-facing applications.
Recommendation: Organizations that have been living in a data warehousing world and supporting analytical/batch workloads need to reevaluate their tech stack with an eye to transactional processing for real-time use cases. They will also need to lean into responsible AI to ensure they don’t cross the line from useful hyper-personalization into being creepy as they interact with their customers.
2. Increasing AI regulation puts spotlight on tools supporting ethical AI
The EU AI Act. The American Data Privacy and Protection Act. The Securing Open Source Software Act. The number of proposed regulations around artificial intelligence is rising rapidly, signaling that the days of companies self-policing their AI/ML projects (or not policing them at all) are coming to an end. Gartner predicts that by 2025 regulations will force companies to focus on AI ethics, transparency and privacy.
Recommendation: Companies must ensure that they have the enterprise model management tools in place to meet new regulatory requirements, like “algorithm design evaluations” and “algorithm impact assessments.” That means being able to track and report on how models were created, trained, tested, deployed, monitored and managed. And if you don’t have an ethics council in place to oversee AI/ML, the time to establish one is now, before regulators come knocking on your door.
3. Model Management becomes the center of gravity for machine learning
ML tools remain highly fragmented. One of the challenges this creates is that, to date, diverse stakeholders (practitioners, management, risk and IT) have not had a common, unified view of the end-to-end ML life cycle. But now, model management platforms will offer a center of gravity tying together the separate tooling around experimentation, production, ML data and orchestration. In effect, a model management platform becomes the control tower from which enterprises will manage all their AI/ML.
Recommendation: Take a holistic approach to machine learning, and consider how you are managing models across the model life cycle. Does your ML tooling have a center of gravity that lets you manage models end to end? Identify and fill gaps in your tools for experimentation, production, ML data, and orchestration, but also implement a model management platform that gives you visibility and control over the entire model life cycle.[…]
Read more: www.forbes.com