When to Open a New Role vs. Outsource to a Robot

Whether you’re a regular ChatGPT user or still on the fence, AI is a fact of life for modern hiring teams. 

According to research from McKinsey, 92% of companies plan to increase investment in AI over the next three years. By 2030, artificial intelligence is expected to create 170 million new jobs—while displacing 92 million more.

Which begs the question: When should you open a new role vs. outsource a set of particular tasks to AI? In this article, we’ll help you understand which tasks and responsibilities can be effectively outsourced to AI and at what point a human needs to step in.

Should you hire a human or outsource to AI? 

As AI tools become the new normal (and vanguards like Shopify CEO Tobias Lütke push for company-wide adoption), it's crucial to know what to delegate to the bots and what to keep human.

The rule of thumb? AI excels at repetitive, data-heavy tasks where human error is costly. But when you need creative thinking or emotional smarts, humans still take the crown.

How to decide which tasks to hand to a human vs. a robot

Ready to divvy up tasks between humans and AI? Let's look at three key strategies to help you make the call.

1. Start with the nature of the work

Start by looking at what the work actually involves. Does it mean crunching numbers or doing the same tasks over and over? That's AI territory. But if you need someone who can think deeply and creatively, you'll want a human at the helm.

As Fast Company's Eric Markowitz puts it:

“The importance of softer skills—like compassion, emotional intelligence, cross-cultural sensitivity, and strategic thinking—is increasing. At the very least, these skills will become ever rarer in a landscape dominated by AI-enabled systems. But AI-driven robots don’t yet have the capacity to interpret complex social dynamics.”

Need a quick way to decide? Carnegie Mellon researchers advise organizations to ask four crucial questions:

• How complex is the task?
• How often does it happen?
• How connected is it to other tasks?
• What happens if something goes wrong?

Think about customer service: Basic call center work? AI can handle that. But a 911 operator dealing with life-or-death calls? That needs the human touch. When the stakes are high, stick with people.

2. Answer the strategy question: bot or brain?

While some tech evangelists want HR to prove why AI can't do a job before approving new headcount, the reality is more nuanced. Even the C-suite isn't immune—one study found 49% of CEOs believe AI could automate most of their role, while only 20% of knowledge workers felt the same.

But don't expect robot bosses anytime soon. As AI and business professor Daniel Tsai points out, “Lateral thinking, creativity, the desire and impulse for innovation, entrepreneurship; those human traits are rather difficult to replicate, at least at this point in time in history, in AI.”

Research backs this up. AI teammates have been shown to actually reduce workplace coordination, communication, and trust.

Before rushing to automate C-suite or knowledge worker roles, consider the following questions:

• What percentage of the role involves repeatable processes vs. strategic thinking?
• How much of the role requires building and maintaining relationships with stakeholders?
• Which specific tasks could be enhanced (not replaced) by AI assistance?
• What unique human qualities (creativity, leadership, emotional intelligence) are essential to success in this position?
• How would automation impact team dynamics and organizational culture?
• What training or upskilling would help the role better leverage AI tools?
• Is there potential for AI to augment decision-making while keeping humans in control?

The key is creating a culture that embraces AI as a tool, not a replacement. Companies that do this well see their teams become nearly 3x more likely to collaborate effectively with AI.

3. Cost-benefit analysis

Everyone says AI is cheaper than human labor. But is it really? Let's run the numbers.

According to MIT research, humans are actually more cost-effective than AI for most jobs. In fact, even computer vision AI (the most advanced kind) can only affordably replace about 23% of workers.

Here's an example of what the math might look like for your average human hire:

Average cost per hire: $4,700

Average salary: $64,000

Benefits: $26,000

Total: ~$95,000/year

For AI solutions, you're looking at:

Custom builds: $10,000-$200,000 upfront

Commercial solutions: Lower upfront costs, but...

Hidden costs: You'll still need skilled humans to run and monitor the AI

Plus, there's what Technovation CEO Tara Chklovski calls the "mind tax" – the mental effort of sifting through AI-generated content and implementing it in real-world workflows.

Before you make the call, ask yourself:

• How much time will we save vs. the money spent?
• Do we have people who can actually run this AI?
• Will we need to hire someone new to manage it?
• How long will it take to get the AI up and running?
• What's the learning curve for our team?
• Can we afford the maintenance and updates?

Bottom line? AI might scale better, but it's rarely a plug-and-play solution. You'll still need human expertise to make it work effectively.

Human or robot: key questions to ask

On the whole, experts agree that AI is often most successful when paired with human expertise – not pitted against it. So how can you decide when and where to apply this new technology?

While it depends on the role and types of tasks, these key questions can help you decide:

• Does this job need the human touch (judgment calls and creative thinking)?
• How many people skills are we talking about?
• Is this task basically paint-by-numbers?
• How often do things get shaken up in this role?
• Is this a people-person position?
• Are we dealing with simple decisions or brain-twisters?
• Will this role help build our company's secret sauce?
• What's the worst that could happen if things go south?
• How much would a slip-up cost us?
• Is this make-or-break for the business?
• Are we looking to shake things up or just keep the trains running?
• Will our customers freak if they find out AI's handling this?
• How much hand-holding are we looking at for training?

When to choose human talent

Hiring humans doesn’t mean avoiding AI – in fact, MIT researchers found that performance increased when humans and AI worked together.

But when making decisions like these, you also need to be aware of AI’s limitations. From hallucinations and bias to data security concerns and legal risks, you should always consult the NIST AI risk management framework before making a decision.

Generally speaking, you'll want actual humans in the driver's seat when you're dealing with:

• People skills (think therapists, coaches, negotiators, healthcare pros)
• Creative heavy lifting (brand storytelling, big-picture planning, innovation)
• Tough calls (crisis management, legal eagles, high-stakes consulting)
• High-stakes stuff (flying planes, medical diagnoses, security, emergency response)
• Cultural savvy (diplomacy, translation, making things work across borders, C-suite leadership)
• Relationship building with clients and partners (sales, customer service)
• Moral compass duties and keeping an eye on things (quality assurance, HR professionals)
• White-glove service (luxury travel, personal styling)
• Face of the company stuff (PR, speaking gigs, political representation)

When to choose AI

"Sure, you can automate pretty much anything if you throw enough time and money at it," as Sizer puts it, "but that doesn't mean you should."

So when does AI make sense? Think: helper bots, number-crunching, coding assists, design support - that kind of thing. Salesforce crunched the numbers and found we can comfortably hand over about half of IT tasks, 40% of ops and customer service stuff, and 30% of finance work to the machines.

Sizer says automation is a good go-to when tasks:

• Run like clockwork
• Follow strict rules
• Need multiple sets of hands
• Are prone to human error
• Can be knocked out quickly

If you're checking off multiple boxes here, you might have found yourself an AI sweet spot.

Getting the best of both worlds

For the majority of roles, combining human oversight with AI tools is the best way forward. Think of it like having the best of both worlds – the efficiency and processing power of AI paired with the nuanced understanding that only humans can bring to the table.

By creating roles that leverage both human soft skills and AI enhancements, you're not just filling positions – you're building a workforce that's ready for whatever comes next. It's about empowering your people to do their best work, using AI as a springboard rather than a replacement.

Whether it's free posting to all the right job boards or easy candidate questionnaires and match scores that screen the right candidates in, Breezy has the user-friendly features you need to make better hiring decisions in a fraction of the time. Because at the end of the day, the goal isn't to replace humans with robots – it's to help humans work smarter, not harder.