The Layoffs That Sparked a Debate
When Microsoft laid off staff at id Software in July, lead services programmer Chris Hays didn't mince words: "They fundamentally don't understand art." The comment sparked immediate debate across the gaming industry, but it reveals something much bigger than gaming politics.
It exposes the fundamental tension between AI automation and creative work — a tension that's playing out right now in customer service.
The assumption? AI can handle the "routine" stuff while humans focus on "creative" work. But this binary thinking misses the real story. Customer service isn't just routine. Every conversation requires creativity, empathy, and problem-solving. The question isn't whether AI can replace creativity. It's whether we're asking the right question at all.
Why "Routine vs. Creative" Is the Wrong Framework
Microsoft's approach to id Software reflects a common misconception: that you can neatly separate routine tasks from creative ones, automate the former, and preserve the latter.
Customer service leaders face this same false choice daily. They're told AI can handle "simple" queries while humans tackle "complex" ones. But anyone who's worked support knows this doesn't reflect reality.
A customer asking "Where's my order?" might seem routine. But what if they're asking for the third time? What if they're a high-value customer on the verge of churning? What if they're asking because they need the product for a wedding tomorrow? The query is simple. The context is everything.
The real breakthrough isn't replacing human creativity with AI. It's augmenting human judgment with AI that understands context.
What Gaming Layoffs Teach Us About AI Implementation
The id Software situation reveals three critical mistakes companies make when implementing AI:
Mistake #1: Treating people as interchangeable with AI. Microsoft apparently assumed that AI tools could simply replace the creative output of experienced developers. Customer service teams face the same risk when leaders view AI as a headcount replacement rather than a capability multiplier.
Mistake #2: Underestimating institutional knowledge. Those laid-off developers carried years of understanding about what makes Doom feel like Doom. Similarly, your best support agents carry deep product knowledge and customer intuition that can't be captured in a training manual. The goal isn't to replace that knowledge — it's to scale it.
Mistake #3: Making AI decisions from the top down. The people closest to the work (like Chris Hays) often see problems leadership misses. Rolling out AI without involving frontline teams leads to solutions that look good in PowerPoint but fail in practice.
How AI Should Actually Work in Customer Service
Here's what we've learned by diving deep into how AI actually performs in customer conversations:
AI excels at pattern recognition across millions of interactions. It can spot a frustrated customer from word choice and sentiment. It can identify account security risks from behavioral signals. It can route conversations based on complexity, customer value, and agent expertise.
But AI also needs to know when it doesn't know. The best AI systems don't try to handle everything — they're built to recognize when human judgment is essential and hand off seamlessly.
This is where the AI-first mindset diverges from AI-only thinking. Leading with AI doesn't mean removing humans. It means redesigning the entire workflow around what each does best.
At Darwin AI, we've seen this play out across hundreds of thousands of conversations. Our AI Workforce handles the high-volume, pattern-based interactions that would overwhelm human teams. But it's designed from day one to collaborate with human agents, not replace them.
The Real Creative Challenge
The creative challenge isn't teaching AI to be more human. It's teaching businesses to think differently about customer conversations.
Most companies still organize support around channels (email team, chat team, phone team) or tiers (Tier 1, Tier 2, Tier 3). This made sense when humans were the bottleneck. But AI changes the equation entirely.
Now you can organize around outcomes. Instead of asking "Which channel did this come through?" you can ask "What's this customer trying to accomplish?" Instead of "Is this Tier 1 or Tier 2?" you can ask "What information do we need to resolve this?"
This is genuinely creative work — rethinking fundamental assumptions about how customer service operates. It requires the kind of institutional knowledge and strategic thinking that Microsoft threw away with those layoffs.
What This Means for Your Team
If you're leading a customer service organization, the id Software story should make you pause.
When you're evaluating AI solutions, are you asking the right questions? Are you looking for tools that eliminate headcount, or capabilities that multiply your team's effectiveness?
The companies that win won't be the ones that automate the most. They'll be the ones that figure out the right division of labor between AI and humans — and that have the humility to keep learning as the technology evolves.
Start by identifying where your team spends time on pattern-based work that AI could handle. Not to eliminate jobs, but to free up human creativity for the interactions that actually require it.
Then look at your AI vendors with a critical eye. Are they selling you automation that reduces costs, or transformation that improves outcomes? There's a massive difference.
The Path Forward
Chris Hays was right — Microsoft fundamentally didn't understand the creative work happening at id Software. But the lesson isn't that AI and creativity are incompatible.
It's that implementing AI requires deep understanding of the work being transformed. It requires involving the people doing the work. It requires humility about what AI can and can't do.
Customer service is creative work. Every conversation is unique. Every customer brings different context, expectations, and needs. AI doesn't eliminate the need for creativity — it raises the bar for what creative customer service looks like.
The question isn't whether to use AI in customer service. It's whether you'll implement it thoughtfully, with deep understanding of what your team actually does, or whether you'll make the same mistake Microsoft made.
Your customers — and your team — deserve better than surface-level automation. They deserve AI that actually understands the work.