BUSI2001 Chap.6 AI and Human-AI Collaboration
AI and Human-AI Collaboration
Generative AI can draft analyses in seconds, which makes the question of who checks them central to business use. This chapter sets out human-AI collaboration as the course frames it: AI contributes speed and pattern-finding while people contribute context, judgement and accountability, and AI recommendations are evaluated by human experts before decisions are made.
It explains the two risks the topic names, cognitive offloading and the empathy gap, and follows the workshop's three rounds on airline data, from human-only Excel work to an AI-only reading of an uploaded file. The worked example shows how an AI insight can be technically true and still misleading when it reports a share instead of a rate.
The practice cards then design the human checkpoints that keep an AI-assisted workflow accountable without losing its speed.
What this chapter covers
- 01
Generative AI to AGI and ASI
- 02
Human-AI collaboration
- 03
Multi-AGI collaboration model
- 04
Cognitive offloading
- 05
The empathy gap
- 06
Human-only, human plus AI, AI-only
- 07
Share versus rate
Worked example · free
Comparing insights from the three workshop rounds
- +1Confirm the share: 3,000 of 3,750 dissatisfied passengers is 80%.
- +1Compute rates within each class: 3,000 / 6,000 = 50.0% and 750 / 4,000 = 18.75%.
- +1Note that the share partly reflects economy's larger size, while the rate shows a genuine gap.
- +1Prompt for a comparison of service scores between satisfied and dissatisfied economy passengers.
Key terms
- Human-AI collaboration
- Working arrangements in which AI supports analysis while people keep judgement and accountability for decisions.
- Empathy gap
- The difficulty AI systems have with tasks that require deep understanding of human emotions.
- Overgeneralisation
- Drawing a broad conclusion about a group or future events from a small number of cases.
- Artificial general intelligence
- A hypothetical AI able to perform the full range of intellectual tasks a human can.
AI and Human-AI Collaboration FAQ
What is cognitive offloading and why is it risky?
Cognitive offloading is leaning on AI so heavily that your own critical thinking weakens. It is risky because it is gradual and comfortable: each unchecked answer makes the next check feel less necessary. Controls include reconciling AI-drafted numbers to source data and occasionally producing analysis unaided to keep skills sharp.
How does the multi-AGI collaboration model keep humans in control?
It combines insights from several AI systems with human expertise and, crucially, sequences them so that AI-driven recommendations are evaluated by human experts before any final decision. That ordering supports trust, accountability and transparency, because a named person remains responsible for what the organisation does.
What should I compare across the three workshop rounds?
Compare the key managerial insights each round produced about dissatisfied passengers: what the human-only Excel analysis found, what prompted Python code added, and what the AI found unaided. Note which insights were new, which were wrong or misleading, and which only a human noticed as relevant to a manager.
Is AGI already being used in business?
The topic presents generative AI as today's tool and artificial general intelligence, and possibly artificial super intelligence, as possible next stages. Business use today centres on generative systems that draft code, text and analysis. The collaboration principles apply either way: humans evaluate recommendations before decisions and keep accountability for the outcome.
Assessment move
Run your own version of the three-round exercise on any small dataset. First analyse it in Excel without AI and write three insights. Then prompt an AI assistant for Python code, run it, and write what changed. Finally upload the data and let the AI interpret it, and audit every claim it makes against your earlier work, paying special attention to shares, rates and sample sizes.
Keep a short log of each error you catch, labelled with the risk it represents: offloading, overgeneralisation or a missing denominator. Practise the five verbs from the topic summary by writing one prompt that questions an AI answer rather than accepting it. These logs make strong portfolio material because they show critical engagement with AI rather than simple use of it.
Write a one-paragraph personal policy for how you will use AI in your assignments: which tasks you will delegate, which you will always check, and how you will record the checking. Revisit and revise it at the end of the semester as part of your final reflection.
Working through AI and Human-AI Collaboration in BUSI2001? Sia is AskSia’s AI Business Analytics tutor — ask any BUSI2001 AI and Human-AI Collaboration question and get a clear, step-by-step explanation grounded in how BUSI2001 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.