Minding the gap: the LBS Data Science and AI initiative

Artificial-intelligence evangelists brandish forecasts so vast they make the Industrial Revolution seem quaint by comparison. Advisory firms have been quick to herald productivity gains, with PwC estimating AI will add $15.7 trillion to annual global GDP by 2030 – more than today’s entire economy of India^[i], while McKinsey & Co triumphantly posits that generative AI alone might contribute $2.6-4.4 trillion annually^[ii]. At the same time, economists such as Nobel Prize winner Daron Acemoglu see it adding less than 1% of GDP^[iii], and overall AI adoption by firms remains sluggish^[iv]. Based on a recent survey, GenAI is used for personal rather than business use, with the most common use being psychological support, meaning and organising one’s life ^[v].

What should we make of these contrasting estimates? There’s no gainsaying that AI, and GenAI in particular, is making bewildering progress. The technology has yet to be broadly adopted, but business uses of AI are developing at break-neck pace and have the potential to transform how businesses add value.

An uncomfortable truth

So, a revolution is unfolding before our eyes. How should firms adjust? To answer this question, we need to address an uncomfortable truth: AI technical brilliance alone won’t magically transform into productivity, profitability or inclusive growth. Real-world gains hinge on the messy human processes of curating data, reshaping workflows, retraining staff and navigating regulatory hurdles.

Navigating this complex intersection of technology, strategy and policy requires rigorous, independent interdisciplinary research – which is precisely why London Business School (LBS) is establishing the Data Science & AI Institute. It will bring every core business school discipline together under one roof to study and shape AI’s impact on business. The mission is urgent and clear: provide evidence-based insights that help businesses, workers and policymakers translate AI’s promise into tangible benefits for firms, employees and society at large.

We need research that avoids the inescapable temptation of vendors and advisors to furnish optimistic predictions of the future and which documents the opportunities and challenges that AI brings to society. We need a forum that bridges business, policy and academia that can help guide corporate strategy and national policy alike.

Why London Business School?

From its base in London – Europe’s pre-eminent finance and innovation hub – LBS operates on a genuinely global scale, with alumni and research partnerships in more than 150 countries. Its interdisciplinary strength is matched by its proximity to the full spectrum of AI-enabled activity: advanced manufacturing, healthcare, life sciences, government services, creative industries, finance and high-growth tech ventures. London’s dense constellation of venture investors, scale-ups and multinational headquarters gives the Institute a living laboratory; while collaborations across Europe, the Americas, Africa and Asia ensure its influence extends far beyond one region. Crucially, the Institute builds on a substantial knowledge base: more than 30% of LBS faculty already conduct research at the intersection of AI and organisations, so it can move rapidly from rigorous analysis to actionable intelligence for companies, workers and policymakers worldwide.

New winners and losers

Consider the nuanced impacts that business school research on AI has uncovered. In a 5,000-agent US call centre, introducing an AI helper boosted productivity 14% overall – but rookies saw gains of 34% while seasoned agents gained nothing; a subtle finding illuminated by business school scholarship on technology and experience^[vi]. The results of a well-publicised experiment involving 735 BCG consultants, half of whom had access to ChatGPT and half did not, showed remarkable productivity gains for those using AI for a creative task, with the benefits accruing to those who would struggle without AI, suggesting AI acts as an “exoskeleton” that supports the weakest performers^[vii].

But the same study showed that relying on GenAI can mislead in settings where advisors need to analyze a complex strategic question, because the prima facie plausibility of what GenAI offers can belie strategic mistakes. Then again, a field-study analysis of a ChatGPT-powered “mentor” showed significant earnings increases for Kenya’s savviest entrepreneurs, leaving less-skilled peers behind^[viii]. Recent research involving software developers using GitHub Copilot highlighted that experienced developers benefitted disproportionately, significantly accelerating their coding tasks^[ix]. AI produces new winners and losers.

Business school research is providing evidence on how technology combines with other factors to reshape business. A study that LBS did with the UK Institute of Directors^[x] shows that senior business leaders believe that GenAI will disrupt some sectors and business models more than others, emphasising the role of regulation, modularity and proprietary data^[xi].

Uneven redistribution

These examples underscore a critical insight: AI is less a simple tool for automation than a force for reallocation. It redistributes productivity and opportunity unevenly; working well in some settings and badly elsewhere – to the benefit of some users and the detriment of others^[xii].

There is also a subtle dynamic at play in how AI reshapes employment. Historically, innovations that lowered costs often expanded demand, creating new roles and markets. Research into ATMs showed how they expanded rather than replaced bank jobs^[xiii], for example. Similarly, early mechanisation grew rather than shrank worker participation in textile industries; not least because lower costs led to increased demand^[xiv]. Today, AI-driven reductions in software development costs do not necessarily eliminate programmer jobs. Instead, they open new markets that were previously economically unviable^[xv]. These second-order effects, essential to grasping AI’s true economic impact, have been brought to light primarily through rigorous research done in business schools.

At the organisational level, business school scholarship emphasises that capturing AI’s value rarely depends solely on technical prowess. Instead, firms thrive by leveraging complementary assets – proprietary datasets, trusted brands, regulatory expertise and distribution networks – that researchers have identified as critical moats transforming fleeting algorithmic advantages into sustained economic gains^[xvi].

Increasingly, studies conducted in business schools demonstrate the effectiveness of integrated “AI factories” – systems that continually train models, deploy solutions and refine processes^[xvii]. Business school research has also clarified when vertical integration, expert-driven solutions and horizontal, employee-led experimentation yield superior outcomes; pinpointing precisely when centralised strategy trumps decentralised innovation^[xviii].

Furthermore, business school studies – including pioneering work by our own faculty – highlight that, for firms to master AI integration, industries must collectively orchestrate it. This research shows that the adoption of AI depends on shared standards, interoperable systems and coordinated investments across supply chains^[xix]. Through detailed ecosystem studies, including the AI ecosystem that differs between major powers such as the US, China and the EU, business school researchers have documented how industries organising these complementarities reap outsized productivity gains, enhancing competitive positioning globally^[xx]. They have also considered the forces that underpin the political economy and geopolitics of AI regulation.^[xxi]

Three pillars

Recognising this opportunity, the LBS Institute for Data Science & AI will rest on three pillars:

Research excellence

Uniting experts across business school disciplines – including strategy, finance, economics, marketing, operations and organisational behavior – the Institute will produce robust research examining the implications of AI for businesses and illuminating its impact at every level: workers, firms, industries and regions.

Industry collaboration and policy engagement

Acting as a neutral convener, the Institute will ensure research insights stand the test of real-world application by engaging directly with tech innovators, firms across diverse sectors and regulatory bodies. It will also help catalyse academic knowledge by offering a forum to bridge practitioners and leading researchers.

Educational innovation

The Institute will pioneer AI-driven educational tools, from generative tutors to data-driven simulations, preparing business leaders to strategically, ethically, and adaptively deploy AI. It will also help identify the skills business school students need to compete in a world where GenAI redefines the very purpose of management education^[xxii].

Crucially, the Institute eschews rigid blueprints. The epicentre of AI shifts with every technological breakthrough and regulatory update. The mandate is not prediction but measurement – continual, rigorous, public evaluation so businesses and policymakers can steer decisively through the AI revolution.

History clearly shows us that productivity revolutions – from steam to electricity to IT – have hinged not on invention alone, but on adoption and diffusion. Silicon Valley might birth the latest chips and code, but prosperity emerges only in the organisations and regions skilled in their application. By illuminating this path from prototype to profit, the LBS Institute for Data Science & AI will ensure Europe is not merely keeping pace but setting it.

References and endnotes

[i] PwC (2017). “Sizing the prize: What’s the real value of AI for your business and how can you capitalise?”

[ii] McKinsey & Company (2023). “The economic potential of generative AI: The next productivity frontier.”

[iii] Acemoglu D. 2024. “The Simple Macroeconomics of AI”. National Bureau of Economic Research, Cambridge, MA: w32487. Available at: http://www.nber.org/papers/w32487.pdf

[iv] McElheran K et al. 2024. “AI adoption in America: Who, what, and where”, Journal of Economics & Management Strategy 33(2): 375–415.

[v] Zao-Sanders, M (2025) “How People Are Really Using Gen AI in 2025”, Harvard Business Review, online, April 9.

[vi] Brynjolfsson E, Li D, Raymond L. 2025. “Generative AI at Work”. The Quarterly Journal of Economics : qjae044.

[vii] Dell’Acqua F et al. 2023. “Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality”. SSRN Electronic Journal. Available at: www.ssrn.com/abstract=4573321

[viii] Kremer, M., Mansour, S., & Perry, J. (2022). “ChatGPT as an Entrepreneurial Mentor: Field Experiment Evidence from Kenya”, Haas School of Business Working Paper.

[ix] Ziegler, J., et al. (2023). “Productivity Assessment of GitHub Copilot,” arXiv preprint arXiv:2302.06590.

[x] Jacobides MG, Ma MD. 2024. “Assessing the expected impact of generative AI on the UK competitive landscape”, Policy Paper, May, Institute of Directors (IoD) & London Business School.

[xi] Stuart ET. 2024. “Could Gen AI End Incumbent Firms’ Competitive Advantage?”, Harvard Business Review.

[xii] Lakhani KR, Yerramilli-Rao B, Corwin J, Li Y. 2024. “Strategy in an Era of Abundant Expertise”, Harvard Business Review.

[xiii] Autor, D. H. (2015). “Why Are There Still So Many Jobs? The History and Future of Workplace Automation,” Journal of Economic Perspectives.

[xiv] Bessen, J. (2015). “Learning by Doing: The Real Connection between Innovation, Wages, and Wealth”, Yale University Press.

[xv] Noy, S., & Zhang, W. (2023). “Experimental Evidence on the Productivity Effects of Generative AI,” MIT Economics Working Paper.

[xvi] Teece, D. J. (2018). “Profiting from Innovation in the Digital Economy: Standards, Complementary Assets, and Business Models in the Wireless World,” Research Policy.

[xvii] Huang, M.-H., & Rust, R. T. (2021). “Engaged to a Robot? The Role of AI in Service”, Journal of Service Research.

[xviii] Adner, R. (2017). “Ecosystem as Structure: An Actionable Construct for Strategy,” Journal of Management; Jacobides, M.G., Cennamo, C., & Gawer, A. (2018). “Towards a Theory of Ecosystems,” Strategic Management Journal.

[xix] Iansiti, M., & Lakhani, K. R. (2020). “Competing in the Age of AI: Strategy and Leadership When Algorithms and Networks Run the World”, Harvard Business Review Press.

[xx] Jacobides M.G., Brusoni S, Candelon F. 2021. “The Evolutionary Dynamics of the Artificial Intelligence Ecosystem”. Strategy Science 6(4): 412–435.

[xxi] Jacobides, M.G., Gawer, A., Lang, N Zuluaga Martínez, D, 2025, “The Political Economy and Geopolitics of AI Regulation”, Management and Business Review.

[xxii] Csaszar, F, Jacobides, M.G, Zemsky, P, 2025, “The Effects of Artificial Intelligence on Management Education”, Strategic Organization, forthcoming.