Business School & GMAT: My Bold Predictions

It’s been a while since I’ve written a long article. Today, I want to share my predictions for the future of business schools and the GMAT exam. GMAC’s research reports on the future of business school employers and the latest data from front-line alumni point to two key trends:

1. Traditional business job roles are shrinking. Data Analyst positions are likely to be hit first, with traditional Marketing and HR roles also in the danger zone.
2. Employers and business schools are increasingly focusing on “how to collaborate with AI” and “AI-related capabilities” as essential skills.

Looking back at the current standardized tests (GMAT/GRE), to be honest, their direct relevance to AI capabilities is not very high. Standardized tests risk falling behind. If they don’t reform significantly and simply become shorter and easier, as some have recently, they’ll become irrelevant.

Let’s take the GMAT, which I’m more familiar with. Even if we remove Sentence Correction (SC) and the Analytical Writing Assessment (AWA), let’s review the remaining sections:

Skills with the Highest Relevance:

• Data Sufficiency (DS) -> This skill is vital for providing sufficient data in prompts when collaborating with AI.
• Critical Reasoning (CR) -> Crucial for fact-checking AI-generated reasoning that might be overly optimistic or pessimistic.
• Multi-Source Reasoning (MSR), Graphics Interpretation (GT) -> These skills are needed to cross-reference large datasets and verify if AI output is accurate or fabricated.

In my opinion, skills with relatively lower relevance:

• Reading Comprehension (RC) -> AI can easily summarize and answer questions about long passages.
• Problem Solving (PS) -> Modern AI, coupled with tool integration, should handle complex calculations without issue.

Therefore, I boldly predict that standardized tests like the GMAT will definitely undergo another wave of changes, with assessments shifting towards “scenarios involving collaboration with AI.”

Given the nature of standardized tests, open-ended assessments like “live AI operation” or “live prompt writing” are unlikely. However, test questions might evolve to look something like this:

You are the Chief Strategy Officer for a mid-sized regional airline. High-speed rail and emerging low-cost carriers have significantly eroded your core business traveler market. Your objective is to use generative AI to conceptualize a next-generation loyalty program that effectively enhances existing business traveler loyalty and Customer Lifetime Value.

Below are several candidate prompts submitted to an AI assistant. Considering the depth and innovation of the strategy, which prompt is most likely to guide the AI in generating a groundbreaking and actionable solution?

A) Act as a market intelligence analyst. Compile loyalty programs from top global airlines and hotel chains, dissecting their membership tiers, reward mechanisms, and partnership strategies. Extract a competitive landscape report, summarizing the top five core strategies most favored by business travelers, and assess their replicability in the regional airline market.

B) Act as a top strategy consultant in the airline industry. For a regional airline facing intense competition, devise a next-generation loyalty program that moves beyond traditional mileage accumulation, specifically for efficiency-focused business travelers. The program should focus on providing “ultimate convenience” and a “seamless travel experience.” Propose three concrete solutions and explain how each embodies the core appeal of “time value.”

C) Act as a user experience design expert. To enhance engagement and retention among business travelers, design a comprehensive gamified loyalty system for our company’s mobile app. This system should include specific achievement badges, advanced challenges, and social leaderboard mechanics. Detail how the design logic aligns with the psychological motivations of business travelers to drive sustained brand interaction.

This question imaginatively blends traditional “Data Sufficiency” (DS) and “Critical Reasoning” (CR) skills with current AI capabilities. Or, consider a question like this:

Background:

You are the Marketing Manager for “Urban Beans,” a chain coffee brand. The company is considering launching a new product nationwide next quarter called “Oat-Nitro Cold Brew.” To forecast market response, your team has used an AI forecasting model and provided the following three pieces of information. Please answer the question based on this data.

Data 1: Sample Customer Data Used for AI Model Training

This table shows the customer data subset used to train the AI model. The data comes from a one-week trial sale at the company’s downtown business district branch in City A.

Data 2: AI Model Sales Forecast Report

(Chart) A steep upward-sloping line graph titled “Oat-Nitro Cold Brew National Sales Forecast for the Next Six Months.” The horizontal axis represents months (from Month 1 to Month 6), and the vertical axis represents forecasted sales volume (in tens of thousands of cups). The line rises linearly from 50,000 cups in the first month to 800,000 cups in the sixth month.

(Text Summary) “Our forecasting model indicates that the new product, ‘Oat-Nitro Cold Brew,’ will achieve significant success in the national market. Based on its broad appeal across age groups during the trial sale in City A, and referencing our historical data from the successful launch of ‘Sea Salt Caramel Latte’ in 2018, we project stable growth in total national sales, reaching an impressive 800,000 cups per month within six months, with a 95% confidence interval.”

Question:

As the Marketing Manager, after reviewing all the information provided by the AI, what is the biggest strategic risk when developing the national launch strategy based on this forecast?

A) The AI’s text summary may contain “hallucinations” because the company did not launch a “Sea Salt Caramel Latte” in 2018.

B) The AI model’s forecast is overly optimistic because the training data may suffer from severe sampling bias.

C) The linear growth shown in the chart is unrealistic, as market response typically fluctuates with seasonal factors.

D) Competitors may launch similar products during the same period, affecting actual sales performance.

This question integrates existing Graphics Interpretation and Table Interpretation, along with Multi-Source Reasoning, into an imaginative scenario.

I was one of the first educators in Taiwan specializing in study abroad exams to research AI’s practical applications, integrate AI into my courses, and advocate for students to use it. Working with AI helped me finally understand why I scored a 760 on the GMAT. It turns out the skills businesses need have always been related to AI, but the existing exam structure just couldn’t highlight them. I’ve put more details in this link:

https://dustingmat.com/en/archives/7513

Current exams will likely move in this direction, unless they want to self-destruct by clinging to outdated methods.

I believe standardized tests will not disappear; schools will still need a standard way to evaluate applicants’ potential. Let’s look forward to a next-generation GMAT exam that truly aligns with employer needs—identifying talent capable of orchestrating AI—and genuinely screens for potential individuals.

Related posts:

Dustin’s 4-Step GMAT Timing Guide Saying Goodbye to My Long-Stuck 500 Score: How I Scored 710 with Dustin GMAT Don’t Treat AI as a Slave Anymore! Teach You the Four-Step Mindset for “Human-Machine Collaboration” GMAT Prep: GPT’s Precision Questions & Deep Dives