AI and Testing Podcast: STARWest Keynote

Ever feel like you blink and the tech world's moved on without you? Yeah. Well, buckle up, buttercup. Today's deep dive, we're talking AI and software testing. Specifically how it's going from, huh, that's neat, to didn't we already talk about this like it was normal?

Fast. Like scary fast. Inflection point fast is what I'm picking up from the notes we got here. Oh, right. Got to lay that groundwork. Exactly. Full transparency, folks. Got my hands on some juicy insider intel the Keynote speech from this big old tech conference.

And let me tell you, this ain't no AI 101 snoozefest. This is peeking into the future, like right now. And you know they always save the spiciest stuff for the the Keynote. No doubt, no doubt. So anyway, Jason Arbon kicks things off with this anecdote, right?

So this is a couple of years back, he was hawking AI testing talks like it was timeshares in Florida. Nobody biting. Empty rooms, tumbleweeds. Had to be that guy practically begging folks to pull up a chair.

Though, got to say, that's the tech world in a nutshell. Ain't it the truth? What's niche today is table sakes tomorrow. And in this case, tomorrow's coming in hot. And get this? Our guy, the the Keynote Jason Arbon, he drops this truth bomb, says we're about to see a massive swing, 20% AI -assisted testing to a whopping 80%, like right around the corner.

Yeah, that's not just like a nudge in a new direction. That's a full -on. No, it's a tectonic shift. For sure. But hold your horses. Before you get too far down the AI rabbit hole here, for the folks at home, what even AI -assisted to testing?

Yeah, no good call. So think of it like two main branches, right? Got your AI -augmented testing. This is where AI tools are basically power -ups for your human testers. Okay, so making the already smart folks even more efficient, basically.

100%, less grunt work, more high -level thinking. Then you got your AI -first testing. That's where things get really spicy. Oh yeah, this is where AI is driving the bus, yeah. You got it. AI's right in the test, running the test.

Humans are mostly checking the receipts, handling the edge cases, the truly- weird stuff. So instead of us teaching AI, it's almost like AI is becoming the developer and the testers are becoming the like the editors.

No, there's an interesting thought. And honestly, not that far off, that shift, AI first, it's coming faster than most folks realize. Okay. Okay. So let me play devil's advocate for a sec. Someone's listening to this, right?

And they're like, cool story, but I'm not building the next Google over here. I get it. But that's the thing, right? This isn't just some like far off Silicon Valley fantasy. This is your banking app.

That shopping app that seems to know what you want before you do. Exactly. That social media platform that might be, you know. We've all got our vices, no judgment. Point is all that stuff built and tested using the same processes.

So AI revolution and testing, that's a revolution in how basically everything digital in your life gets made. Oh, yeah. That hits different when you put it like that. It's not just tech, it's life. You know, I got to admit, when I first wrapped my head around AI testing, I'm picturing like typos, broken links, you know, the robot stuff.

The stuff even humans hate doing. Exactly. But then I'm like, how's AI going to tell me if an app is, you know, nice to use if it feels good? It's the question, right? Like can a robot have good taste?

And honestly, that's what's blowing my mind about this new AI. It's getting good at the human stuff. Okay, man, you got to back that up and give me an example. So the Keynote mentions AI giving feedback on user experience.

Not just like, can your finger hit the button? But is that button where a human brain expects it? Stuff we used to argue about in design meetings for hours. No more whiteboard battles. Hey, maybe AI will take our jobs eventually, but for now, less time arguing, more time shipping good stuff, I'll take it.

And get this, they're using AI to like, roleplay different users. Okay, now that sounds like my kind of video game. Right. But instead of like, level 50 elf mage, it's how is a cyber security expert going to use this?

Or what if your user needs accessibility features, testing for way more than just, does it crash? So it's almost like building in better software from the get go, not just finding bugs, but making it good, smarter, you know, for everyone.

Bingo. And here's the kicker, faster, cheaper, because AI doesn't need sleep, lunch breaks, or to complain about its stock options. Music does some manager's ears, I'm sure. But okay, we've geeked out on the how enough.

Let's talk why. AI testing, why should anyone care? Besides sounding cool, which it does. Gotta admit. Look, I'm a cynic at heart, too much hype in this industry, but this, this is way more than that.

the Keynote dives deep into the proof. Real world results happening now. First off, not just catching the little stuff, we're talking AI finding vulnerabilities, the kind of stuff that used to cost companies a fortune.

And now it's like AI flags it, human testers swoop in like, we got this. They become the cleanup crew, the elite squad, not stepping the weeds all day. But it's bigger than bugs. Jason Arbon talked about performance analysis.

to be. You test your app speed, you get a number, done. AI, it's comparing you to the competition in real time. So like our app loads 10% faster, bam, and you can prove it. And you know, companies pay big bucks for that kind of intel.

And then standardization. This is huge. AI runs the same tests on everything. Suddenly it's apples to apples. No more OR processes, different excuses. So on, AI brings this like objective truth to the table, no more fudge in it.

And that's before we even get to the future stuff. Oh, there's more. the Keynote painted this picture, right? Testing isn't just pass fail anymore. It's granular, it's nuanced, it's understanding how your software really stacks up in the real world.

So it's not just does it work, it's how good is it, really? And here's where my brain broke a little. AI generated code gets big, right? He's saying we'll have this loop. AI writes the code, AI tests the code, AI uses that data to write even better code.

So like, does anyone even have a job at that point? Asking for a friend. He addressed that head -on. And here's the good news. Testers, developers, we're not obsolete. We're just leveling up. Okay, love a good power -up.

Think less button pushing, more strategy. We become the conductors, AI's the orchestra. We interpret the data, we find the patterns, we tell the AI where to dig deeper. So like, AI's the hammer, we're the architects.

And the really good architects are gonna be in high demand, let me tell ya. But look, we've thrown a lot at ya today. AI upending software testing, doing things we thought only humans could do, the benefits are real, it's happening.

But for you, dear listener, what's the takeaway? Well, Jason Arbon ended on this. Adopt .ai, yesterday. Don't wait for the future to run you over, get ahead of it. If you're still manually testing something AI could do faster, cheaper, well, sounds like it's time to have a conversation with your boss, or your budget, whichever listens better.

And this ain't about replacing humans, it's about us using tech to be better humans at our jobs. More efficient, more effective, ultimately more competitive. Which, let's be real, in the tech world, that's how you survive.

Now Jason Arbon ended on this kind of mind blow idea. AI levels the playing field. Small startups, they can suddenly compete with the big dogs on quality, even without the same resources. Imagine, software quality, not limited by your budget, just by your imagination.

That's the future he's painting, and honestly, I'm here for it. As always, thanks for joining us for this deep dive. Until next time, stay curious, stay informed, and keep on exploring the wild world of tech.