Twitxy Saved

Rob Butler 🇨🇦

Verified account This account is verified. Learn more

@RobButler

Harvard Data Science Agentic AI
ProSci Change Management
Government of Canada | #GCDigital | Views are mine.
Canada ca.linkedin.com/in/robertcbutl… Joined February 2009
5.1K Following 10K Followers 25K Posts

Latest posts 5

Rob Butler 🇨🇦 Verified account @RobButler
Reducing red tape is not the same as reducing governance

TBS is running a reset of the Policy on Service and Digital. One of its four stated aims is better outcomes by reducing red tape and focusing on results.

I want to sit with that phrase for a minute, because it does a lot of work and it can be read two ways.

The first reading is subtraction. Fewer requirements, fewer gates, fewer forms. Faster by definition.

The second is harder. Most of the process we experience as red tape was written in response to something going wrong. Remove the requirement without replacing the assurance it was providing, and you haven't reduced burden — you've moved risk somewhere it isn't being watched.

The distinction that seems to matter is between process that produces assurance and process that produces paperwork. Both feel identical from inside a delivery team. Both eat the same weeks. But only one of them is doing a job.

Sorting those apart is slow, unglamorous work. It needs people who understand why a given control exists, which is often not written down anywhere and lives with someone two reorganisations away.

The engagement numbers suggest they're taking that seriously: just under 70 sessions across departments, all provinces and territories, and 11 governance committees since July 2025. Drafting is underway.

If you sat in one of those sessions — I'd be interested in what you argued should stay.

Modernizing the Policy on Service and Digital, TBS: canada.ca/en/government/…

Views my own.
#GCDigital #PublicService
Rob Butler 🇨🇦 Verified account @RobButler
The one sentence that changes the AI conversation

TBS published the Guide on the Use of Agentic Artificial Intelligence in May. Buried in the overview is a sentence that should reframe how we all talk about this:

Generative AI may describe what should be done. Agentic AI can attempt to do it.

That is not a technical footnote. It is a governance dividing line.
Here is the part I think gets missed. Agentic AI does not swap one set of risks for another. It inherits every generative concern — accuracy, privacy, fairness, security — and then adds a second set on top.

The new set is about actions rather than content. What did it touch? What permissions did it have? Can we see what it did? Can we undo it? Most of us are far less practised at these, because until recently we did not need to be.

So the honest framing is not "agentic AI is the next thing to assess." It is "agentic AI is everything we were already assessing, plus a category we have barely started on."

The guide's answer is two principles layered on top of FASTER. Bounded autonomy, meaning tight explicit limits on data, tools and scope, with read-only as the default. And recoverability, meaning the thing can be paused, stopped and returned to a safe state, with every action logged somewhere the agent cannot edit.

Both of those are design decisions, not procurement decisions. They have to be made before anyone builds anything, by people who understand the program. Which means the governance conversation cannot wait until there is a tool to assess.

If your department is planning agentic pilots this year, the question is not whether the AI is good. It is whether you can explain, in one sentence, what it is allowed to do and what happens when it does something you did not expect.

Can you?

Guide on the Use of Agentic Artificial Intelligence, TBS: lnkd.in/gTDH663Q

Views my own.
#GCDigital #AIGovernance #PublicService