Episode 74: Building an Information Management Foundation for Human-AI Collaboration

Tori Miller Liu04/03/2024

The rise of artificial intelligence, especially large language models like ChatGPT, has organizations buzzing about the transformative potential of technology in the future of work. Amidst these new advancements, data sits in the center—powering how organizations can fully leverage these tools for their unique business needs.

In the latest episode of the #shifthappens podcast, AvePoint Chief Brand Officer Dux Raymond Sy, along with AvePoint Chief Marketing & Chief Strategy Officer Mario Carvajal, welcomed Tori Miller Liu, President and CEO of the Association for Intelligent Information Management (AIIM) to explore the vital connection between AI and information management.

Liu shared her expert perspective on the shifts AI drives, the importance of data readiness, and the steps companies can take to foster an innovative, human-centric culture amidst technological change.

The Inevitability of Change and Shining a Light on Information Management Gaps

In this episode, we piloted a new segment called "The Sound of Shifts," where we ask our guests to share a song that comes to mind when they think of shifts happening. When asked what song reminds her of the current AI-driven shift, Liu pointed to Bob Dylan's iconic "The Times They Are A-Changin'." The classic lyrics, she explained, capture "the inevitability of change" happening in society and technology.

"I feel like at this moment in time, from a societal and technology perspective, that's what we're experiencing," said Tori. "There's quite a lot of inevitability around AI — the pace that it's accelerating, the pace of adoption. And hopefully for the better and to improve lives. But if you don't move with that change and educate yourself, you will be left behind to a degree."

One of the core themes Tori emphasized is how AI has "shined a spotlight" on existing information management issues that organizations previously could ignore or put off addressing.

"We've had the luxury of ignoring unstructured data for quite some time," she said, referring to information like emails, contracts, schematics, and other documents outside of structured databases. "When talking about generative AI, I don't think you can afford to ignore that data anymore. You need high-quality data to make AI work and make it sustainable."

Dux supported this by noting that companies have been through similar cycles with new technologies like cloud computing. The promise of finally getting data organized and tagged properly was there but often kicked down the road.

"If we rewind ten years ago when the cloud was a thing, everybody moved to the cloud, and this was the talk track - 'We're moving to the cloud, we'll clean up, we'll organize and classify and tag,'" recalled Dux. "And then what happens ten years later? You move to a new house, but your boxes are still there."

AI has made ignoring poor data quality and information management an untenable position. Dux also recounted a story of a customer who, when using Microsoft's AI copilot, could surface sensitive files they didn't realize they had access to simply by tweaking the prompts.

"Nobody's being malicious...but in an accidental way, it may be out there," said Dux. "So we need to put guardrails in place to automate a lot of this enforcement to ensure data quality and security."

The Need for Universal AI Governance

Another key issue Liu stressed is the necessity of universal standards and frameworks for AI governance to mitigate risks and provide clarity for responsible development. She cited the EU's AI Act as a promising model emphasizing risk mitigation over strict control.

"If folks want to get involved in the regulations around AI, now would be a great time to write your representatives and talk about the importance of a universal framework," urged Liu. "We don't want a repeat of GDPR, where it was confusing and challenging for enterprises to comply with that regulation. And then every other country and state came up with their own flavor."

Liu also argued that AI developers should be proactive in areas like data traceability, noting potential future requirements to store training data could have significant logistical implications. "It's important to pay attention to now because it's going to have infrastructure repercussions for organizations if it comes to fruition."

Building Logistics & Culture for AI Success

So, what should companies be doing today to prepare? Liu broke it down into three key areas:

1. Focus on the "Why": According to Tori, AI is a tool just like any other technology that we've been exposed to in the last 30-40 years, and success is dependent on people and their understanding of the importance of the initiative—it's all about change leadership and change management.

1. Prioritize Data Logistics: Cross-functional conversations must happen around accessibility, interoperability, security, and data quality—the building blocks for AI success.

1. Plan for Sustainable Innovation: Organizations should think about operationalizing and scaling AI so it's safe for the business, employees, and consumers. Tori also emphasized the need for new metrics of success around innovation that include scalability, safety, and improving human lives.

However, these logistics should be developed to improve the human experience, according to Mario and echoed by Liu.

This entails rethinking technical integration and the entire digital workplace infrastructure — from collaboration tools to virtual workspaces. How can AI and future technologies like artificial general intelligence (AGI) integrate into these environments intuitively and empoweringly?

"I expect the systems to know who I was talking to and where my last edits were," Mario said. "If we can bring all this together, maybe we can get rid of the mundane tasks and focus on each other — being more productive, changing things, and supporting customers better."

Fostering an Agile, Innovative Culture

Fundamentally, Liu stressed that developing analytics skills and fostering a culture of continuous learning will be critical for long-term AI adoption and workforce success.

"It's worth noting that there are applications of AI that not only introduce operational efficiencies but can impact your workforce," said Liu. "I think it's an opportunity for HR departments and executives to start figuring out how to upskill or reskill their staff. If it's not possible, how do I find other opportunities for them?"

She advocated nurturing characteristics like agility, innovation, and a willingness to experiment in workplace cultures. After all, the future impacts of AGI and other advanced AI systems are still unknown.

"I don't know how I would prepare my staff to think about 'Hey, at some point, AGI might do 50% of tasks,'" Liu said. "Maybe it's better to focus your culture on agility, innovation, and not being afraid to fail or experiment."

Laying the Groundwork for AI with a Solid Data Foundation

The road ahead for responsible AI adoption has many complexities to navigate. But as Tori imparted, taking proactive steps around data readiness, governance frameworks, and cultural philosophies can help organizations be prepared.

"I hope when people are listening to this, they don't feel so overwhelmed that we're saying, 'You need to stop and look at your information management strategy, data quality, interoperability, and security,'" said Liu. "There's an entire ecosystem of consultants and solution providers that can help make this manageable and break it down for organizations of any size."

Establishing a solid foundation today can clear the path for humans and AI systems to collaborate for better experiences tomorrow. As Bob Dylan would say, "The times they are a-changin' - and the opportunities are there for the taking."