It’s funny. For a company that has spent the last two-plus years covering artificial intelligence from every possible angle, we were actually pretty conservative about using it ourselves.
When ChatGPT burst onto the scene in late 2023, we did what everyone did. We played with it. We asked it to write poems. Summaries. Jokes. The usual parlor tricks that made people who hadn’t seen AI before lean back in their chairs and go, “Whoa.” It was impressive, sure. But it wasn’t something you trusted with real work. Not if you cared about accuracy, tone, or reputation.
Our internal posture at Techstrong was simple: Interesting tool, not ready for official output.
Then, sometime last year, it got real.
The models improved. A lot. Not just in cleverness, but in usefulness. ChatGPT, Claude, Gemini and even Grok for a while started producing things that were actually usable in a business context. At the same time, something else happened that doesn’t get talked about enough. We got better at using them. Prompting stopped being guesswork and started becoming a skill.
Yes, there is slop out there. Tons of it. But here’s the other side of that coin. AI has enabled an explosion of output. Just as developers are shipping dramatically more code thanks to AI assistance, we are publishing dramatically more content. That’s not a coincidence. The volume of news around AI itself has been growing at warp speed. Without these tools, we simply could not keep up.
For a while, though, agents were the exception.
When agentic AI started getting buzz in late 2024, our first reaction was blunt: These things kind of sucked. They were hard to configure, fragile, and raised legitimate security concerns. Most of the time, getting them to do something useful required more effort than doing the task yourself. We reported on the space, but internally, there was no rush to deploy.
Then came the moment that changed everything.
Much like when Netscape lit the fuse on the web, a new generation of tools suddenly made agents feel accessible. Call it OpenClaw or whatever label you prefer. The name isn’t the point. The experience was. People who tried it started saying the same thing: “I feel like I can build anything.”
At Techstrong, it began with a challenge. I pushed one of our more technically savvy producers to give it a try. He resisted at first. Security implications alone were enough to give anyone pause. But after some prodding and careful investigation, he installed an instance on a Mac mini.
That single machine triggered a tsunami.
Within days, multiple team members were running their own local instances on their own Mac minis. We learned how to lock them down. How to manage permissions. How to connect them to the tools we use every day. Most importantly, they started doing real jobs.
Not demos. Not experiments. Work.
Agents began reading transcripts from our shows, analyzing videos, drafting content, editing copy, publishing posts, scheduling updates, summarizing agendas and coordinating tasks. They were watching material that would take a human hours to review and producing structured outputs in minutes. The effect wasn’t that people stopped working. It was that they suddenly had time to do higher-value work.
Then came the next wave.
Perplexity introduced its Computer capability as part of a premium subscription. Two hundred dollars a month is not pocket change. But I asked a simple question: Compared to the fully loaded monthly cost of an employee, what does that price really represent?
I made an offer to anyone at Techstrong who wanted in. The company would pay for one month. In return, participants had to give me weekly updates on exactly what the system was doing for them. At the end of the month, we would decide together whether the investment made sense.
It’s still early. But the results so far have been extraordinary.
We’ve automated major portions of newsletter production. Video descriptions and social posts are being generated at scale. Articles are moving through publishing workflows faster. Agents are helping design and deploy web pages. They summarize our daily Techstrong Gang agendas before meetings even begin. And this is after only a few days of serious use.
We’re also exploring bigger operational changes. Learning events. Sales outreach. Internal processes that haven’t been questioned in years are suddenly on the table. Personally, I’ve been using agents to help manage email, triage calendars and streamline workflows that used to eat entire chunks of the day.
This is not about replacing people. It’s about amplifying them.
Our goal is not layoffs. It’s multiplication. Give talented professionals leverage and watch what happens. A producer who can accomplish ten hours of work in two suddenly has eight hours to innovate instead of grind.
That said, I’m not going to pretend everyone is thrilled.
Some people are resisting. That’s human nature. Change is uncomfortable, especially when it feels like it threatens the skills that made you successful. But leadership requires honesty. If your role can be dramatically enhanced by these tools and you refuse to use them, that becomes a problem. Not because we want fewer people, but because the world outside our walls is moving just as fast.
The most striking example of this shift came from a conversation with one of our editors.
He was adamant that AI would never be able to publish articles directly to WordPress. The formatting alone, he argued, would make it impractical. We debated. That’s part of our culture. I asked him to try anyway.
Later that afternoon, I walked into his office and asked how it was going.
He looked up and said, almost casually, “I just reinvented our entire editing and publishing process.”
That’s the kind of report I keep hearing.
We’re not just experimenting anymore. We’re redesigning how work gets done.
For a media company, that’s profound. Our business runs on speed, accuracy and volume. If agents can handle the repetitive mechanics, humans can focus on judgment, storytelling and strategy. The parts that actually differentiate us.
In the tech industry, we like to talk about “eating your own dog food.” Using your own products before asking customers to trust them. What’s happening at Techstrong goes beyond that. We’re not just eating the dog food. We’re howling at the moon.
If our experience is any indication, the next few months are going to bring changes that make the last two years look slow. Entire workflows will disappear. New roles will emerge. Organizations that adapt will surge ahead. Those who don’t will struggle to understand what happened.
The future isn’t some distant abstraction anymore. It’s sitting on your desk, logged into your tools, waiting for instructions.
My advice is simple. Don’t fear it. Don’t ignore it. Learn it. Use it. Stretch it until it breaks and then learn how to fix it. The people who do that will become dramatically more valuable, not less.
Opportunities this obvious don’t come along often.
If you see the future standing in front of you and choose to look the other way, you shouldn’t be surprised when it walks past you and doesn’t come back.
