Five Things: May 28, 2026

NOTE: I’m pretty public about my “AI” skepticism, but there are a handful of people I trust to have informed, nuanced takes that help challenge and refine my own thinking. (Two are featured below.) I’ve also had several private conversations with friends and colleagues about their own usage, which I appreciate because most of them are well aware of my skepticism. I’ve still yet to come across any compelling tools or use cases that are of interest to me, but I actually do continue to look for examples that might change my mind. YMMV.

---

_ONE

Writing & Publishing Awards Have Difficult Decisions to Make Regarding AI | Jane Friedman

While prohibiting AI might be appropriate for some, enforcing such a policy (if it is a real policy and not merely a wish) requires effort and resources that will have to be continually upgraded and assessed. How hard do you want to fight, and how much time and money do you want to commit? I can imagine a couple paths forward depending on the resources of the organization.

The list of AI LLM-related fiascos grows longer every day, but the “inevitability” narrative continues. As Friedman notes here — covering one of the more interesting high-profile situations in publishing this year (so far); not that one. nope, not that one, either. this one, which I hadn’t even heard about until now! — it is definitely inevitable that companies and organizations are being forced to establish clear guidelines and policies for how they’re dealing with “AI”, even when their preference is to not engage with it at all.

Simply making a statement about not using (or not allowing the use of) “AI” is meaningless without a way to enforce and police that goal, especially in any scenario where you’re collarborating with people beyond your control. “AI” tools have been shoved into virtually everything we use on a daily basis, and many companies don’t have clear guidelines or training for how employees should (or shouldn’t) use them, never mind freelancers and other external partners.

Perhaps unsurprisingly, Microcosm is ahead of the curve: “Friend, we know you can do better.” Most publishers or organizations who are legitimately concerned about “AI” usage could easily adopt a similarly transparent and nuanced policy, although I suspect most won’t go that far.

__TWO

Human Driven Entirely Loopy | Dominic Wellington

Applying AI in isolation might look like productivity, at least at the level of the individual, but the overall result is not improved by nearly the same margin, or perhaps at all.

I’ve come to appreciate Wellington’s clear-headed perspective on technology, particularly his ability to discuss “AI” in a broader historical context. His blog is a recent addition to my blogroll after a strong run of insightful posts like this one, and he’s become one of my favorite follows on LinkedIn, in particular, where he’s one of the only people to get me to watch a video.

I’ve had similar experiences being given allegedly ‘good enough’ “AI” outputs that were, at their best, bloated and clumsy approximations of intelligent thinking, far from ready for prime time viewing. At their worst, I’ve suffered from second-hand embarrassment, and a loss of trust and respect in people who are “ideating” and “collaborating” with their preferred tool(s) of choice, and then publicly sharing the results as finished products.

This “content” is almost always produced by someone without any notable expertise in the topic they’re focused on, typically looking for a shortcut or attention. (Or, way too often, both.) Combined with automated graphic design or generating entire presentations, you have a “human in the loop” recipe for making someone else’s job even harder while you’re bragging on LinkedIn about how you’ve 10xd yourself. #cmonson

___THREE

AI search is creating a new incentive system for media | Pete Pachal

All this suggests that AI systems are as ‘gameable’ as search engines and social networks, just in a different way… AI systems reward retrievable substance, not necessarily the most insightful or information-dense content.

Similar to SEO, I believe we’re going to see two paths emerge with “AI” search: good actors optimizing technical structure while prioritizing writing for humans, and bad actors optimizing technical structure to game the system by writing for the algorithms. Again. It’s a sequel no one asked for, with a larger budget and a script that’s being written on the fly — and too many people gullibly pretending the ending will be different this time.

The publishers and marketers who gamed the system in the past benefitted from inflated traffic and valuations, and almost all of them are now shadows of their former selves; zombie brands acquired for pennies on the dollar, desperately hoping for an AI resurrection.

The publishers and marketers who stayed focused on serving actual humans definitely struggled as search and social traffic took a hit, but those were mostly glancing blows because they prioritized building solid foundations on direct relationships with readers. That has always been, and still remains, a winning approach — regardless of the latest new shiny.

____FOUR

In Defense of Theory | Kerry Cunningham

The narrative that MQLs drive revenue growth isn’t just a metric preference. It’s load-bearing folklore. Challenging it doesn’t feel like a methodological disagreement. It feels like an attack on the company’s past success.

I recently came across Cunningham via Andrew Mitrak’s insightful A History of Marketing podcast, thanks to his episode’s provocative title: “The MQL Industrial Complex & Where B2B Marketing Went Wrong.” It’s rare to see B2B marketing get the spotlight outside of B2B channels, and I was intrigued by the idea that something might be wrong with MQLs, a concept I’ve never been a big fan of anyway.

Despite a lot of my career being in B2B marketing and lead gen, I’ve mostly avoided dealing with MQLs as an explicit metric. While the underlying concept makes sense, particularly alongside SQLs, I’ve been fortunate to have avoided going too far down the marketing automation rabbit hole for them to ever be relevant to me. I’ve also rarely had to deal with sales operations that prioritized and measured reps on trackable, automatable activities.

Hearing Cunningham’s revelation that MQLs have always been flawed and his proposed alternatives was enlightening and heartening, and I immediately started following him on LinkedIn, where he posted this manifesto. It’s a rarity on LinkedIn: a long, insightful read with receipts and solutions, and better yet, isn’t actively trying to sell you anything.

My one caveat is his focus is primarily on larger B2B operations, where five-figure deals and robust CRMs are the norm. If that’s your world, it’s a must-read. If it’s not, it’s still a great read for anyone in B2B marketing or sales, because the fundamentals still apply.

_____FIVE

Too Many Tables, Too Little Progress | Michael Cairns

The people most qualified to lead these organizations are spread so thin that meaningful strategic direction and guidance become nearly impossible to sustain. And in an environment where resources are increasingly strained, leaders spend most of their productive time managing membership subscriptions and sponsorships. It is certain that large publishing players are reducing their financial commitments across the board making the financials for many associations even more constrained.

I’ve been thinking about this a lot recently in a specific context (IYKYK), but Cairns does a great job outlining the bigger picture for the entire publishing industry, and offering some logical solutions to a clearly defined problem. The alphabet soup of organizations is unsustainable, many with overlapping and occasionally conflicting agendas. All too often, the best resourced ones have the worst priorities, while the ones with the least resources need to be “charming at dinner” to keep their small staffs employed.

When I was running Panorama Project several years ago, the problem was particularly glaring because it was an initiative that arguably should have been a BISG effort instead, but 6.5 years later, publishers would still rather gouge libraries and promote misinformation about them [that’s a Substack link, unfortunately; an extremely rare exception here] rather than collaborate with them on reasonable and mutually beneficial solutions.

Related, I’ve been pleasantly surprised by how well the IBPA / PubWest merger has played out so far, and really enjoyed (their Executive Director) Andrea Fleck-Nisbet’s authentic enthusiasm about what the combined organization is doing on her recent BookSmarts appearance.

______BONUS