Last week, I wrote about the challenges enterprises have when implementing AI like traditional IT projects, including their frequent struggle to show value, tendency to go over budget, and failure to achieve true workflow transformation. I argued that instead of IT-led AI projects, the real action is happening with worker-led AI use on the edge, where individual workers are using LLM-based platforms like ChatGPT.
People reacted to what I wrote by saying, “Ah, so the enterprise AI issue is really about BYO” or “shadow AI.” On the surface, yes, today’s worker-led AI has many similarities with the consumerization of IT we’ve been talking about for decades. But reducing worker-led AI to a “just a BYO problem” is the wrong way to think about it and creates blind spots that will hold companies back.
While companies are spending millions trying to find the benefits of AI, rank-and-file workers have already found it, getting real results by incorporating AI and redesigning their workflows based on what these AI platforms can do today. This is so much more than shadow AI!
If a company reduces all that worker effort down to a compliance box-check, they’ll end up with the worst of both worlds: approved locked-down tools that no one uses and a workforce quietly reinventing workflows without them.
As I wrote previously, ChatGPT, Claude, Gemini and other consumer AI tools are no longer just chatbots. They’re extensible platforms with connectors, context windows, and integrations that can drive genuine workflow change. Workers are following a clear progression as they adopt these tools—from simple copy-and-paste helpers to ambient copilots to agents that can actually operate browsers & computers.
This is so much more than when “BYO” meant playing around with an unapproved app or trying to get your work email on the latest phone. Modern workers are using consumer AI tools to prototype the future of your workflows on the fly. Every time someone wires up a prompt chain, or connects their assistant to a SaaS tool, or figures out how to automate a chunk of their job, that’s an experiment in what tomorrow’s work could look like.
Using terms like “BYO” or “shadow” to describe workers using their own AI tools is nice because it lets us classify and bucket our response. Unfortunately, those terms also hide and minimize the real value workers are creating here.
So instead, let’s reframe this worker-led AI as capability adoption at the edge. This framing helps us see goals in the context of broader business impact and not just security risk. It also lets us reframe how worker-led AI can interface with the business:
Your enterprise is already structured around humans. I recently explained why letting AI operate inside the same human-shaped environments is the most practical, secure, and scalable way forward. If the AI tools driving these environments come from the outside, then that’s where the focus will need to be. Treating the workspace as the control plane where apps, identity, security, and context converge will support both today’s hybrid world (humans and their AI tools) and tomorrow’s evolving mix.
What workers are figuring out with their LLM-based tools is not a nuisance. It’s the innovation engine companies say they want. A million micro-experiments are already running, prototyping the next generation of workflows. Blocking that frustrate employees and starves your own R&D.
Worker-led AI isn’t shadow IT. It’s shadow strategy. Treat it that way: secure it, enable it, connect it, and scale it. The companies who do will capture the competitive advantage currently being discovered and figured out at the edge. The ones that don’t will be left managing compliance checklists while their workforce quietly works around them and writes them out of their own org chart.
Read more & connect
Join the conversation and discuss this post on LinkedIn. You can find all my posts on my author page (or via RSS).
My upcoming public talks
Brian Madden is a VP & futurist at Citrix. He writes about the future of work, AI in the workplace, and the evolution of Citrix.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。
