When I joined SAS in 1997, most analytics workflows still revolved around desktops, batch processing and highly technical users.
Later that same year, SAS introduced SAS/IntrNet – a technology that helped bring SAS analytics into the growing world of web applications. At the time, it felt like a major shift in how people would access data and insights.
Nearly 30 years later, I’m watching another shift unfold.
This time, the conversation is about generative AI, AI agents and autonomous systems that can reason, act and adapt in real time.
During a fireside chat at SAS Innovate 2026, Mel Robbins said something that stuck with me: change feels fast today, but it will never move this slowly again.
That line immediately made me think about how dramatically both technology and the role of SAS users have changed since my first year here.
What made SAS/IntrNet such a big shift?
When SAS/IntrNet launched in 1997, the internet was rapidly changing how businesses thought about software and information access.
At the time, analytics environments were often decentralized and highly technical. SAS/IntrNet helped change that by allowing organizations to bring SAS analytics into web-based environments, making insights more accessible beyond technical teams.
-
- Exterior view of the former SAS Denver office in 1997, where the internal SAS/IntrNet kickoff took place. Photo courtesy of fellow SAS employee Gerry Nelson.
-
- One of the office views from Gerry Nelson’s workspace at the SAS Denver office during the 1997 SAS/IntrNet era.
-
- Another view overlooking the area surrounding the SAS Denver office in 1997, captured from Gerry Nelson’s office during the SAS/IntrNet kickoff period.
For SAS users, it also marked a shift in how we worked. Programmers who traditionally focused on analytical workflows suddenly found themselves helping build web-enabled applications and broader enterprise experiences.
Looking back, it felt like SAS was embracing a major technology shift happening across the industry.
From information delivery to autonomous intelligence
Today, I see strong parallels between that era and the AI conversations happening now.
In 1997, organizations were trying to understand how the internet would reshape access to analytics and information. In 2026, organizations are asking how AI systems will reshape decision-making itself.
Businesses no longer want systems that simply generate reports or summarize historical data. Increasingly, they want intelligent systems that can retrieve context, automate workflows and help support decisions in real time.
That’s where generative AI, AI agents and agentic AI become especially important.
The focus is shifting from AI that simply responds to prompts toward systems that can reason, act and adapt within business operations.
From SAS/IntrNet to agentic AI
| 1997 | 2026 |
|---|---|
| Web-enabled analytics | Autonomous AI systems |
| Information delivery | Decision orchestration |
| Desktop workflows | Embedded intelligence |
| Technical specialists | Cross-functional AI governance |
| Static outputs | Real-time adaptive systems |
How the role of SAS users has evolved
One of the biggest changes I’ve witnessed over the last three decades is the evolution of the role of analytics professionals alongside the technology itself.
In the late 1990s, much of the value centered around technical implementation – writing code, building analytical processes and delivering outputs.
Today, those technical foundations still matter, but the conversation now includes governance, oversight and ensuring AI systems operate responsibly within real-world environments.
In many ways, the role has evolved from simply producing analytics to orchestrating intelligent systems.
Why trust and governance matter more than ever
As AI systems move closer to execution rather than simple assistance, trust becomes even more important.
Questions organizations are now facing
How are AI decisions governed?
What safeguards exist when systems act autonomously?
How do organizations balance speed with accountability?
What role should humans continue to play in oversight?
How do businesses build trust in AI-driven operations?
Those questions matter because AI is becoming more operational and embedded into real business workflows.
That’s one reason this current era feels so significant to me. The technology is evolving rapidly, but the need for trusted analytics and responsible innovation remains constant.
What SAS/IntrNet and agentic AI have in common
Although the technologies are very different, I believe SAS/IntrNet in 1997 and agentic AI in 2026 share an important similarity: both changed how people interact with analytics.
SAS/IntrNet helped move analytics beyond isolated desktop environments and into broader organizational workflows through the web. Agentic AI is now helping move intelligence beyond static outputs toward systems capable of reasoning and acting within operations.
Both moments also raised important questions around accessibility, governance and trust as technology became more deeply embedded into how organizations operate.
Reflecting on nearly 30 years at SAS
I’ve personally witnessed almost three decades of technology evolution at SAS and across the broader marketplace.
From the rise of web-enabled analytics to today’s conversations around agentic AI, the pace of innovation continues accelerating.
At the same time, this reflection reminds me how important it is to appreciate each chapter along the way.
I’d especially like to thank my colleague Allan Manning for teaching me SAS/IntrNet back in 1997 and another colleague, David Weik, for helping me better understand today’s world of agentic AI. Between those two moments are countless colleagues who helped shape my career and understanding of technology across multiple eras of innovation.
As SAS celebrates its 50th anniversary, I feel grateful to have experienced these transformations firsthand and even more curious about what comes next.
