Welcome to Complex Machinery!

If someone pointed you here and you just want to subscribe, you can do that right here.

(Be sure to click on the confirmation link in your e-mail.)

If you're looking for more details, read on:

What's this all about?

Complex Machinery is a newsletter about AI and risk. It occasionally dips into related topics like complexity, business models, emerging tech, and marketplaces.

While the newsletter covers AI, there are no technical deep-dives. This is higher-level, matters-to-business-leaders thinking.

That feels like a grab-bag of topics

It may look that way from a certain perspective, but those topics are all closely related:

• Risk and complexity go hand-in-hand. (Complexity leads to all kinds of incidents)
• AI is my field of expertise, so I'm very much in-tune with the risk/reward tradeoffs
• I started my career in finance, which has a lot of parallels to The Field We Now Call AI (and used to call predictive analytics, data science, then machine learning, then AI, and now AI again because generative): it's all about analyzing data for fun and profit
• I am something of a broken record because I keep pointing out how the rise of computers on Wall Street will tell us a ton about how AI will impact other fields. You will see mention of this in pretty much every other newsletter.

"Risk?" Isn't that just about bad stuff happening?

The term "risk" has several definitions, which change based on the context. Some definitions focus on what goes wrong, but really, risk is about upside as well as downside.

Tell me more about the format

Complex Machinery lands two or three times a month. Issues run anywhere from 1200-2000 words.

Some issues are a grab-bag of short segments, while others are essays.

Who's behind this newsletter?

Complex Machinery is written by Q McCallum. You can read more about me on my website but the gist is: I've spent my career in and around the emerging-tech space, and I've been in (The Field We Currently Call) AI since the early days of "predictive analytics" and "Big Data."

These days my work sits at the intersection of risk and AI. This newsletter is a space for me to share my thoughts.

So are you some kind of AI crank? A skeptic?

As I mentioned, I work in the AI field. As part of my job, I help companies surface use cases, perform due diligence, and that kind of thing.

Given that:

• I see the true potential for AI and have high hopes for what it can achieve.
• I point out when what is advertised doesn't quite align with reality.

That doesn't make me a crank or a skeptic. Just an experienced practitioner.

Aha. So this newsletter is your marketing vehicle.

Only in the sense that what you see here reflects what I think about the AI space.

I occasionally mention other projects I'm working on, but that's about it.

If you'd like to retain my consulting services – imagine the kind of thinking that goes into the newsletter, but applied to your specific business challenges and concerns – you're welcome to contact me.

But if you're looking for a typical business-style newsletter, which is really a thinly-veiled brochure that lands in your inbox every couple of weeks … I don't have that.

I'm almost sold. Before I subscribe, do you have a list of highlights I can browse?

Some of my favorites include:

• #002 - "It's still a wild animal": There's a common problem with LLMs: a little something I call The Random™.
• #018 - "Robot jobs": Sussing out when AI is a good fit for a task, using the humble self-service kiosk as a guide
• #033 - "GenAI is an executive's best friend": Why executives love AI, and why they probably want to pull back a bit..
• #022 - "Does this count?": Can we trust when a computer says that something has happened? Maybe … but we should have a think. And leave room for doubt.
• #010 - "The looming AI debt wall": There's the question of whether AI can ever live up to its promises. Will it make good before that debt comes due?
• #013 - "Blue screen for Armageddon" and #014 - "Embers and ashes": A two-part analysis of the CrowdStrike incident, which was a harsh lesson in risk and complexity. When everything's connected, one small disturbance can have widespread impact.
• #017 - "Stacking the deck": My recurring theme of "the rise of robots on Wall Street tells us a ton about what'll happen when AI enters other fields."
• #029 - "Setting it in motion": What history teaches us about giving things away for free, and how that relates to the news about DeepSeek giving away its "R1" model.

If that sounds interesting to you, you can subscribe here.

(Be sure to click on the confirmation link in your e-mail.)

What's your privacy policy?

The privacy policy is simple: I only use your e-mail address to send you the newsletter.

I don't share or sell your data, nor any other euphemism marketers employ for "showing little respect for your contact information."

Where else can I find your work?

Complex Machinery – here – is the best spot. If you like what I have to say, that's the space that is updated most frequently.

Still, if you can't get enough of my thoughts, you can check out:

• my blog (not updated as much, but the archives go back a long ways)
• My O'Reilly Radar articles – longer reads on topics as varied as remote work, N-sided marketplaces, automated machine learning, and navigating the pandemic.
• My one-pager websites, such as "Will AI Help Here?" and "How Do I Do AI?"
• I occasionally post on Bluesky at @qethanm.bsky.social
• If you prefer to follow people on LinkedIn, I also have a modest presence there.

I also have a couple of other projects in the works. I'll mention those as they land.

I'd like to contact you about other matters

For professional (consulting) matters – working with your executive on AI strategy, use cases, and risk management – I'll point you to my website.

For general inquiries, that site also has a contact form.

I'm interested. How do I read Complex Machinery?

You're welcome to browse the archives online, or subscribe below to get the newsletter in your inbox.

(Be sure to click on the confirmation link in your e-mail. If you don't see it within a few minutes, it's probably in your spam folder.)