How Much Does AI Matter for Print?

How Much Does AI Matter for Print?

By Noel Ward, Editor@Large and AI Skeptic

I don’t know the answer to the question in this title but you’re the pros, so you tell me. I’ve been to some print-related conferences and the AI sessions are always well-attended. Yet when I ask people afterwards (one-on-one) responses range from skeptical to cautiously optimistic. No one seems to be betting their business on AI.

Thank goodness!

Most use AI as a tool to do a few things faster but seem to be more or less experimenting. This is good because they learn both their limits and those of AI. Some of the most data-centric printers I know only test it, doing what I think of as “sanity checks.” Some use AI as an aid for decision-making—seeing it as one tool of several. Others use it to fine-tune language when responding to an RFP. All are smart guys with significant operations who recognize a useful tool when they see it but limit AI use to an advisory capacity, another voice in the room, if you will.

Given that, I want to hear what some Print Planet readers are doing. Please tell me a couple of things:


You can respond in the comments section or email me privately at [email protected]. All responses sent to my email are confidential and I will not identify anyone who responds, either in the comments or via email.

Thanks for your help. We all get smarter with more input from colleagues.

The number of views and responses is enlightening.

gordo said:

The number of views and responses is enlightening.

Click to expand...

I think it's because we are all scrambling to try to keep up with it and the output in a 'real world' vs 'theoretical world' environment isn't predictable or stable enough yet. I feel like the winners of this race in the next 5 years are going to be those who took the time to make sure they know what's going on with AI but at the same time, I don't have the time and energy to keep up with how it works AND you know... do my actual job. So.. the bored paper pushers who aren't actively being paid to sit around and interact with AI are going to end up being the winners.

My 2 Cents.
It's gonna be useful but right now only when I'm not busy doing things.

Diddleysquat.

IMHO - the departments that will most likely be hit or replaced by AI in implementation order:

1 Estimating
2 Scheduling
3 CSRs
4 Preflighting
5 Prepress
6 Sales
7 Pressroom

Here's my problem (and the reason I previously gave the one word "diddleysquat" answer).

All AI systems (Large Language Models), are "taught" by humans, who are, by very definition of being "human", fraught with errors, mistakes, imperfections, and biases. Which is why almost everything generated by AI comes with some sort of disclaimer such as "Generated with AI, which may contain mistakes".

In a lot of businesses, a "mistake" simply means temporary embarrassments, egg on face, ooops, my bad, etc.

On the other hand, a mistake in the print and mail business could cost the owner tens, if not hundreds of thousands of dollars in reprints and postage reimbursements over the course of a year.

If I were not retired, and, still in the business, I'd be verrrryyyy wary about incorporating AI into my business.

"Posting Generated with AI, which may contain mistakes". LOL