Software Isn’t Dead: AI Agents, SaaS Panic, and the Interface Shift

This episode of AI First with Adam and Andy is an intentional experiment. The script, ideas, and analysis are entirely Adam Brotman and Andy Sack’s own. The on-screen hosts and voice delivery were generated using AI tools as part of a hands-on exploration of what these technologies can and cannot yet do. It is not perfect, and that is precisely the point.

In this conversation, Adam and Andy tackle the increasingly popular claim that “software is dead.” Against the backdrop of roughly $830 billion in software market value repricing, they unpack the rise of AI agents, the pressure on seat-based SaaS models, and the market’s tendency toward one-sentence apocalypse narratives.

They explain why software is not disappearing but evolving, as the interface shifts from dashboards and logins to ask, decide, execute workflows layered over systems of record. For restaurant and retail operators, this shift has real implications for permissions, audit trails, orchestration layers, and enterprise control.

The core message is clear: software is not dying, but the way leaders interact with it is changing.

What is AI First with Adam and Andy?

AI First with Adam and Andy: Inspiring Business Leaders to Make AI First Moves is a dynamic podcast focused on the unprecedented potential of AI and how business leaders can harness it to transform their companies. Each episode dives into real-world examples of AI deployments, the "holy shit" moments where AI changes everything, and the steps leaders need to take to stay ahead. It’s bold, actionable, and emphasizes the exponential acceleration of AI, inspiring CEOs to make AI-first moves before they fall behind.

Forum3 (00:00)
This is AI First with Adam and Andy, the show that takes you straight to the front lines of AI innovation in business. I'm Andy Sack and alongside my co-host, Adam Brotman. Each episode, we bring you candid conversations with business leaders transitioning their businesses with AI. No fluff, just real talk, actionable use cases and insights for you.

Today's episode is a little different and I want to be really clear about that. This episode was generated using AI.

Adam and I worked together to draft the script and we use Synthesia to create the host you're seeing and hearing. The ideas and perspectives are ours. The delivery is AI generated. We're sharing this because we're actively experimenting with these tools. We're curious about what's possible right now and we think it's important to be transparent with you about how we're using them. All right, so the take that's been everywhere the last couple of days is quote, software is dead. And the market is kind of acting like it wants to believe it.

So let's just slow down. What are we actually seeing here and what's real versus what's just a good headline?

Yeah, if you zoom out, you can see it in the category. This isn't one company getting punished for a bad quarter. It's the whole software and cloud universe repricing at the same time. And the market is trying to jam it into a one sentence story, which is AI agents are coming and seat based SaaS gets complicated.

gets weird is a nice way to put it.

Yeah, gets weird, gets scary. And by the way, I think part of the reason it feels so intense is because agents is one of those words that everyone uses differently. Is a chatbot an agent? Is it a workflow? Is it a system that can actually take actions? It's confusing even for people in the bubble. But the narrative is doing what it always does. It takes a real trend and then it sprints past reality into apocalypse mode.

Yeah. And the move is not small. I think I saw that the other day, the S and P software and services group dropping almost 4 % in a session. Then it keeps sliding the next day. And the part that made me go, okay, this is not normal is the market cap math. And as of recording this, something like $830 billion wiped out since Jan 28. That's not vibes. That's fear. Yeah, that number is insane. And what's wild is how fast the market decides

There's exactly one explanation. Like, it must be agents. Agents are replacing everything. It's a neat story. It's also usually wrong in the details. Exactly. It's not the databases vanish. It's that the value moves up the stack. The agent becomes the interface, the decision maker, and a lot of the old SAS UI starts to feel like plumbing. Yes, and I think that's the right way to say it.

It's not that software disappears, it's that the interface shifts. Humans spend less of their day living inside the UI. And that's why the seat-based model freaks people out. Investors are imagining fewer seats, fewer logins, less time and app, less reason to pay for the whole suite. If an agent can take a plain English instruction and run the workflow across systems, you start asking, what am I paying for? The UI?

or the outcome.

They love one sentence stories. And Oracle is a great example of how extreme it got. They had what this brutal stretch and the framing was basically A is going to replace enterprise software. Plus, why are you spending so much on AI and is it going to pay off? And those two things together are like gasoline.

Totally. And I'm not picking on Oracle, just using it as a clean example of how these narratives stack. You get replacement fear plus capex skepticism at the same time, and that combo is just brutal. And then everyone acts like there's no middle ground, when in reality, there's always a middle ground.

And what's interesting is the smartest people are split. You've got one camp doing full apocalypse framing, like this is the end of sass. And then you've got the other camp basically saying, guys, calm down.

Yeah, and I'm sympathetic to the calm down camp. Sanofsky calling the software apocalypse is thing nonsense. I get it. The arm CEO calling it micro hysteria also kind of nails the vibe. And Jensen's point is the one people skip past. AI doesn't float in space. It runs on software. So software doesn't disappear. The question is where the value accrues, who captures it and how it's priced.

And for restaurant and retail operators like our customers, this is not theoretical. Nobody wants 40 dashboards. Nobody wants their district manager spending an hour screenshotting reports into an email. But also, you still need systems of record. You still need controls. You still need audit trails. Like, if labor is blowing up at three stores, you need to know who changed the schedule and why, and you need it to be defensible.

Totally. Systems of record aren't going anywhere. If anything, they matter more because the agent needs something trustworthy underneath it. What changes is how humans interact with those systems. It becomes an action layer on top. Read, decide, execute with permissioning, approvals, and audit trails. That's the shift. Okay, but bring this down to earth. Because when a restaurant exec hears, you won't be using software.

They're like, what are you talking about? Walk me through it. What does a Tuesday look like?

Yeah, when I say that, don't mean software disappears. I mean you're not spending your day jumping between five tabs and 40 dashboards. The interface starts to look like the places you already are. Text, email, Slack, WhatsApp, eventually voice. You talk to it like you talk to a person. And under the hood, it's pulling from your data and hitting your existing systems. And I've been seeing these early agent systems that actually do it through messaging. OpenClaw is one example.

And I'm telling you, it is both fascinating and terrifying. Fascinating because it feels like the first real glimpse of the interface shift. Terrifying because it's like, wait, this thing can actually do stuff in the real world on real accounts. And to be clear, a lot of this is still super early and not at all secure at this point. But you can see where it's all heading. To put it in your terms, this is a real holy shit moment.

Yes, I totally know what you mean. I've been playing with Claude Cowork recently and it's that same feeling. The moment it stops feeling like a chat bot and starts feeling like honestly, a coworker who can take a task and run with it. It's like, ⁓ this is different. And you get that little adrenaline spike where you're like, this is amazing. And then immediately you're like, wait, this is also insane. Exactly. That is the exact emotional arc. You go, this is the future.

and then two seconds later you go, this is how things go off the rails.

If you give it the wrong permissions, if you hook it up to the wrong thing, if someone compromises one account, now you've got an agent that can move fast and it doesn't get tired. It doesn't get bored. It doesn't go home at five.

Totally, guardrails are the whole game. Permissions, least privilege, human in the loop for the stuff that matters, and really thoughtful logging. But just sit in the operator experience for a second. Most COOs are not like, I can't wait to use more software today. They're like, tell me what matters, tell me what changed, and help me fix it. Let me make it really concrete, because this is where people's brains snap it into focus. It's 7, 18 a.m.

You're not at a laptop. You're in the car or you're grabbing coffee or you're walking the floor. And instead of thinking, okay, I need to log into four or five different platforms, then email three people. You just pick up your phone and text your agent like you text your DM. Like, give me the 12 stores I should worry about today. Rank them. What changed since yesterday? And what's the driver, labor, comps, staffing gaps, food cost? And tell me what you're not sure about.

and it responds with a clean list. Store, issue magnitude, why it thinks it's happening and what it recommends. Not a dashboard screenshot. The actual so what. Then you keep going. Draft a note to each GM with the top three issues, keep it short and make it specific. Open the tickets that need to be opened. If scheduling is the problem, propose a fix that stays inside labor targets, but don't

push changes live until I approve. And the key is, it doesn't just suggest, it can actually do the workflow across your systems. It can create the ticket, tag the right person, attach the context, propose a schedule adjustment, and then stop at the permission boundary. You get an approval screen that says, here's what I'm about to do. Here's the impact. Here's the audit trail. And the difference in operator life is enormous.

You're not spending your morning assembling information. You're delegating action with guardrails. Yes. Or like, hey, why did chicken cost spike this week in these regions? Or which promo is driving the weirdest substitution behavior? That's the stuff people actually want, not another dashboard.

Exactly, and that's the crux of what the market is trying to price, not software dies, but where the control point moves. If the day-to-day interface becomes ask, decide, execute, then the value shifts toward whoever owns the orchestration layer, the permissions model, the integrations, the workflow engine, the identity and audit trail, the layer that can safely route work across systems and prove what happened. The prettiest UI.

matters less if nobody wants to live in the UI. The winning products feel less like a place you go and more like a capability that shows up where you already are and can act on your behalf. Inside rules.

And that's the scary part for incumbents. If your mode is people are trained to click your buttons, that mode gets thinner. If your mode is you're the system of record with deep workflows and real controls, you still matter maybe more than ever. Exactly. There is so much more we could talk about here and start to bring the audience along with some of our latest ideas. But let's save that for another show. Yes, for sure. For now, the takeaway.

Software isn't dead, but the way we touch is gonna change. Thank you all for listening to AI First with Adam and Andy. For more resources on how to become AI First, you can go to our website, forum3.com. Download case studies, research briefings, executive summaries, and join our email list. Also, we want to invite you to connect with us on our AI First community, a curated hub and network for leaders turning AI hype into action.

We truly believe you can't over invest in your AI learning. Onward.