beckarahn@gmail.com

The era of “just let the AI do whatever it wants” may not last very long.

We asked our software engineering team a question: “Given how fast AI changes, how has your job changed in the past 6 months?” Their answers suggested that software engineers need to be a little like time travelers with a crystal ball: figuring out how to apply tomorrow’s capabilities to today’s problems.  Just a few months ago, AI-generated code still felt unreliable. Developers would use it to code quickly, but with the expectation that they’d have to come back later and fix everything themselves. When something broke, they dove into the details to repair it. Now, the role is becoming less about writing code and more about directing systems. Instead of fixing problems line by line, developers explain what’s wrong, point to evidence, and let AI handle the implementation. The skill is moving up a level—from doing the work to guiding it. Less typing, more directing. Less fixing, more diagnosing. Even routine tasks reflect this shift. Rather than manually enforcing standards, developers can teach AI to do it for them: run the right tools, follow the right rules, and [...]

There’s a growing idea in the AI-driven tech world that quality can come from sheer volume. Instead of carefully crafting one solution, you generate dozens—maybe hundreds—of versions, test them automatically, and pick the best. Build an app, rebuild it 20 times, run AI tests on each version, and then someone chooses the version that works best. And that last part is really important: someone chooses the version that works best. Even if creation and testing are automated, judgment isn’t. Someone still has to evaluate the results—and that requires context, not just output. You can see why this is important by looking at manufacturing. Many experienced workers are retiring, and with them goes decades of hard-earned, unwritten knowledge. Newer workers may know how to operate machines, but they don’t know the history behind them—why a part fails a certain way, which sounds are harmless, or which ones signal serious trouble. Sometimes that knowledge is subtle. A squeak might seem like a problem to fix with a little oil, but it could actually be a built-in warning sign that a [...]

What if incorporating more AI into your business meant that you could invest less of your time, get more value from that time, and then be good humans, be compassionate to yourselves and others in the time and the freedom that that creates? You would essentially be able to give that AI benefit back to your employees to allow them more time to breathe, to deal with external stressors, and allow your employees to do more excellent work.  The 40-hour work week of a typical office worker isn’t really filled with 40 hours of work. The actual important, creative work probably only takes up a fraction of the time that they are sitting in their chair. That’s the work that requires a deep understanding of your business and your customers, and that's the part you need to preserve, nurture, and develop.  Modern Logic’s CEO, Dustin Bruzenak, says, “Years ago, I had a job as a young developer at a really large company. I remember my day would be to show up, boot my computer, do my email, and [...]

For this week’s post, we want to highlight an article by Arvind Narayanan & Sayash Kapoor from Columbia University. We think they have an interesting way of thinking about AI that is very different than the Hollywood-style “machines-taking-over-the-world” storyline.  Their article proposes that AI should be viewed as “normal technology”, a powerful but ordinary general-purpose technology like electricity or the internet. Its economic and social impacts will likely unfold gradually over decades, because invention, commercialization, adoption, and organizational change all take time.  Instead of replacing humans, AI will most likely create a division of labor, with machines automating certain tasks while people supervise, integrate, and make higher-level decisions. Many sci-fi movie plot elements, such as technological accidents, misuse, concentration of power, and geopolitical competition, could also apply to the adoption of earlier new and revolutionary technologies. The authors argue that focusing too heavily on speculative existential scenarios can distract from the practical policy challenges already evident today. Overall, they recommend treating AI as an evolving technology embedded in society rather than a singular transformative event. AI, on the [...]

For this post, we asked our friend Travis Adney to discuss how his business utilizes AI. Travis is the Director of Innovation at Lawrence & Schiller and the Managing Director of their new Minneapolis office. His role involves explicitly identifying where marketing and technology intersect and what unique opportunities he can create for clients, as well as empowering his staff. We asked: What's one of your biggest successes, maybe, or biggest challenges in AI? “There have probably been just as many challenges as there have been big wins, but that's because there's no playbook to this type of thing. My first passionate point is that you need to decentralize AI. Even though AI is on my job description, I can't be the AI guy. We have focused a lot on education with our staff: here's how to use AI well and how to use it wisely. I think you have to continually talk about it. It has to be a conversation every single day. It can't just be something that you schedule in as a one time meeting. [...]

AI-enabled browsers like Dia and Claude have recently introduced the concept of skills. A skill is a reusable set of instructions, scripts, and resources that allows the AI to perform specialized, repetitive tasks. Think of it like a recipe for your AI output. Traditionally, you provide your AI with a prompt and receive an answer. When you teach your AI a skill, instead of one answer, you can set it up with a workflow of tasks to do. For example, a SWOT analysis is a classic business strategic planning tool to identify an organization's Strengths, Weaknesses, Opportunities, and Threats. Instead of creating its own analysis, you can teach your AI to perform a SWOT analysis as a skill, allowing it to automatically format the information in a way that aligns with your business's existing standards. Here’s another great example of a skill. We have set up a workflow to generate outreach and we’ve taught our AI that “outreach” for Modern Logic includes things like emails, text messages, and meeting invites. It knows how to look up information in [...]

AI-enabled web browsers like Dia and Atlas from OpenAI are what’s new in AI this week. What makes them different? They enhance your web browsing experience by adding an AI tool on the side, allowing you to chat in real-time with the web page you're viewing. We’ve talked before about the benefit of adding context to your AI prompts, and these new browser tools are basically doing that step for you. Here’s a great example: We have a customer with numerous Amazon reviews, and they would like to analyze that data. But Amazon doesn't have an export reviews feature. With these new browser tools, you can visit the Amazon review page, open the ChatGPT interface, and ask it to provide a summary of the reviews in a format compatible with an Excel spreadsheet. We’ve all had the experience of searching for a recipe and having to scroll through pages of ads, stock photos, and stories about grandma to find the information you need. This marketing-driven experience may become an artifact of the internet age, fading as these AIs [...]

We’ve heard numerous conversations from experienced developers stating that the new AI capabilities and tools are transforming how they work. But one thing to keep in mind is that there's a lot of coverage of those success stories and a lot of articles extrapolating that for one experienced developer working on their own. The claim is that because AI can bring phenomenal increases in one person's productivity, so therefore, a team of 10 people will see a proportional increase in productivity. However, that might not be entirely true. A Forbes article recently stated that in their survey group, at least 40% of employees received "workslop" in the last month. What is "workslop"? -- that's the term for when AI is used to generate plausible-looking but incorrect output, and a human being passess that work off as their own. It ends up being work that a coworker has to redo. AI is great at creating something that appears, on the surface, to be exactly the deliverable you need. But when you look into it deeper, one problem we commonly [...]