We talked a couple of months ago about how the Front Half of the Hype Cycle is a roller coaster.  As a society, we are constantly getting pumped up about some new technology, overestimating its potential impact, the crashing back to earth.

It’s an exciting roller coaster ride, if somewhat exhausting.

What are some applications in the Front Half right now?  And what are some strategies for addressing technologies currently going through the peaks and valleys?

Applications in the Front Half

In the first article in this series, we talked about three technologies that feel like they are in the Front Half right now.  Let’s dig in a little deeper for some case-in-points:

• Agentic AI
• Quantum computing
• World models (a new term to me just last month)

The Agentic AI buzz during 2025 was overwhelming.  We’ve seen wild, fantastic claims for how Agents could replace white-collar jobs, often as the rational for layoffs.  Hundreds of startups (probably thousands) are pitching Agentic ideas to investors right now.  And now we’re seeing a wave of developments around better security/governance tools to help companies control the leashes around the Agents’ necks – Microsoft’s recent Ignite conference was full of those kinds of announcements, and we’ve seen similar press releases from OpenAI and Anthropic.  Maybe the best sign we could be in a Trough of Disillusionment: there’s a hilarious podcast about a guy who tries to start a company with AI agents as co-founders.  Hilarious, or maybe soul-crushing?

I’ve been fascinated to hear the conversation around Agentic AI shift towards “is this really just Robotic Process Automation (RPA) with a new name”.  There’s good arguments for that mindset, and honestly a smarter, higher ceiling RPA might be a really good outcome for Agentic AI.  We’ve rarely seen RPA get to its full potential at real companies, because it usually requires a lot of business process and programming work to get tools like Power Automate, UiPath, Appian, etc. to really work.  Could adding AI help RPA finally break through?

Quantum computing feels like nuclear fusion – perpetually 10 years out.  IBM in particular has put many billions of dollars into that business, with effectively zero revenue so far.  I spent about 10 hours last month digging into O&G applications for quantum computing (huge thanks to friends Wesley (Wes) Dyk and Bob Parney), and I’m persuaded that quantum computing is still a few more years away from commercial impact.  That said, the potential applications are definitely getting clearer as vendors get more real-world experience with these tools.  Route optimization and complex completion design optimization are areas that look interesting for E&P’s.

World models (or at least the 2020’s version of them) are new enough that they haven’t made much noise in the mainstream technology press.  But they are big news in the entertainment industry for rapid generation of content for video games and movies (World Labs’ Marble, Google’s SIMA 2, Hunyuan, etc.).  Go watch some eye-popping YouTube videos if you’re into that kind of thing.  In the E&P world, it feels like world models might be able to unlock a step change for Digital Twins (whose impact has been underwhelming so far) or oilfield development planning (a notoriously tough nut to crack).  Hold on tight, another roller coast ride is coming.

Screenshots from Hunyuan’s latest stuff. Color me impressed, preferably with some Van Gogh sunflowers.

Strategies for the Front Half

Three approaches will help you get a better ROI from time you spend on cutting-edge technologies:

• read/listen widely
• start with problems (not solutions)
• keep a level head

First, read and listen widely.

As a guy trying to run a business and raise kids and indulge my audiophile tendencies and occasionally hit the gym, I can’t chase every technology rabbit.  I can’t afford to spend 20 hrs comparing Claude to ChatGPT to Gemini or mess around with open-source MCP servers.  What I can do is try to find smart people interested in similar things to what I’m interested in, and follow what they write and speak about.  Increase what you’re aware of, and you’re basically taking more shots on goal for potential impact.

Second, start with problems (not solutions).

One of Amazon’s famous leadership principles is “start with the customer and work backwards,”, and I’ve found this to be immensely true when working with early-stage technologies.  I’ve seen untold hours wasted by trying to shoehorn a cool technology into an application that isn’t a good fit.  (Admittedly I was often the one wasting time.)  Don’t start with a technology and flow downstream to applications, start with problems and fight upstream to technology solutions.  It’s counter-intuitive, but it works.  The broad awareness you get by reading and listening widely is key here.  If you aren’t aware of a technology or its potential applications, your odds of finding something with a big ROI are vanishingly small. “Chance favors the prepared mind.”

Finally, keep a level head.

When you’re in the Front Half, it’s easy to get either overly optimistic as you read a sunshine-pumping article or overly pessimistic as you watch a doomer video.  It’s never as good as the best thing you read, nor as bad as the worst.  Bear in mind also, your content diet (social media, news, etc., etc.) hugely impacts your view of where a technology is on the roller coaster.  If I’m reading and listening to different people than you are, of course I’m going to have a different view on how promising something is!  It’s less stressful to ride the roller coaster when you accept that it will eventually calm down on its own.

Wrap-up

So what do you think?  How do you feel about riding up the Peak of Inflated Expectations or falling into the Trough of Disillusionment?  Does this get you excited to hop on another trip or wear you out?  Give us a call, we love to talk about this kind of stuff!