Generative AI (GenAI) is likely the most significant technological breakthrough of the decade, yet the impacts are hard to quantify. Even detailed studies produce contradictory data, and while the evidence suggests a handful of hours saved per worker per week,1 there’s no clear correlation to revenue. How can we definitively say whether GenAI is worth the investment?
The fast food industry, also known as the Quick Service Restaurant (QSR) industry, can give us a clue. For decades, QSRs have been collecting and measuring data on customer interactions, but only recently have they started to implement GenAI solutions in the drive through with automated order taking.
Fast Food Is a Powerful Litmus Test for GenAI
Automated order taking in QSR drive throughs serves as close analogy to the AI chatbot interactions that enterprise workers are having every day. Both GenAI use cases rely on large language models (LLMs) for Q&A, but the drive through features a well-defined transactional format that’s fast—most drive through service times take about five minutes2—and scaled up to millions of orders daily. Most importantly, there’s money involved.
The endpoint of these interactions has a direct impact on a QSR’s bottom line. The increased efficiency of automated order taking means more transactions, shorter queues, and potentially more upsells during each transaction. These outcomes are far more tangible than trying to measure worker productivity in hours saved. What do the numbers say?
On Paper, GenAI Is a Tremendous Value Enhancer
Most QSRs aren’t publicly releasing the financial data on their GenAI pilot programs, but there’s a treasure trove of speculative and survey data worth looking at.
Industry analysis estimates the initial hardware cost of automated order taking at about US$15,000-20,000, with an additional US$25,000 per year in ongoing software expense, compared to US$60,000 per year in annual labor costs to staff a drive through.3 As labor costs increase and technology costs decrease, the margin will likely only widen.
AI also beats out human workers when it comes to accuracy and speed. Per a recent survey, AI achieved an order accuracy of 95% compared to 89% overall and reduced total service time by 29 seconds.2
Tracking the Direction of Fast Food Giants
With these numbers in mind, you would correctly assume that QSRs are eagerly jumping on the GenAI opportunity. Industry analysts predicted that 2025 would be the tipping point, with major chains like Wendy’s and Taco Bell piloting automated drive throughs across hundreds of locations.4
The outlier is McDonald’s, an early mover on the automated drive through scene. The McDonald’s pilot began in 2021 and ended in 2024, relegated to the back shelf without a full-scale deployment. Per a franchisee memo, a senior vice president remarked, “While there have been successes to date, we feel there is an opportunity to explore voice ordering solutions more broadly.”4
Cracks in the Surface, or Cracks in the Foundation?
It’s safe to assume based on McDonald’s’ hesitancy that the on-paper figures don’t readily translate to the real world, but their communications don’t indicate total abandonment of the technology either. It’s easy to find anecdotal evidence of sensationalized AI mishaps online,5 and this speaks more to how the technology is still maturing as is consumer sentiment. 45% of surveyed customers said they “don’t like the idea” of automated drive throughs,6 but actual customer satisfaction scores reach 98% when AI is involved compared to 94% when it isn’t.2
When you combine these insights with the fact that an AI system may be cheaper than human labor, but the human worker generally completes a lot of other tasks in the kitchen, it makes sense why McDonald’s is pressing pause on the whole endeavor. In short, the technology is promising but not perfect. Panera Bread founder Ron Shaich echoed this point when he said that the winners of automated drive throughs will be “fast followers, not first movers.”4
Full Circle: GenAI Litmus Test Results
So what’s the big picture? The automated drive through experiment provides some measure of the success of GenAI, results best characterized as middling and not breakthrough. However, there’s also an air of inevitability about the GenAI future. Sentiment is catching up to satisfaction. Interest and funding are as strong as ever, with venture capitalists pouring nearly 60% of their investments into AI.7 Aiming to be a “fast follower,” as Shaich put it, implies it’s just a matter of time.
The main takeaway: the rocket is still on the launchpad, but you still need to build the rocket.
- Ars Technica, Time saved by AI offset by new work created, study suggests, May 2025.
- Intouch Insight, 2024 Drive-Thru Study: Key Insights from our Annual Report, Oct 2024.
- QSR Magazine, What’s Next for McDonald’s? U.S. Growth, and Perhaps Some Drive-Thru AI, Nov 2023.
- CNBC, AI drive-thru ordering is on the rise — but it may take years to iron out its flaws, July 2024.
- Business Insider, TikTokers are roasting McDonald’s hilarious drive-thru AI order fails, Feb 2023.
- Intouch Insight, AI in Customer Experience: The Future of Drive-Thru Efficiency, Oct 2024.
- PitchBook, AI eats up 58% of global venture dollars as fear of missing out drives up dealmaking, April 2025.
