Generative Artificial Intelligence in Small and Medium Enterprises: Navigating its Promises and Challenges

The latest technological developments in generative artificial intelligence (GAI) offer powerful capabilities to small and medium enterprises (SMEs), as they facilitate the democratization of both scalability and creativity. Even if they have little technical expertise or financial resources, SMEs can leverage this technology to streamline work processes and unleash innovation, thereby improving their product offerings and long-term competitiveness. This paper discusses how SMEs can navigate both the promises and challenges of GAI and offers a roadmap for deploying GAI. We introduce a sailing metaphor that reveals key strategic dimensions for GAI deployment: competency of employees, effective leadership and work values, organizational culture, collaboration and cooperation, and relationships with third parties. We offer practical recommendations that serve as a useful compass for successfully deploying GAI in SMEs.

💡 Research Summary

This paper investigates how generative artificial intelligence (GAI) can become a strategic lever for small and medium‑size enterprises (SMEs) that typically lack deep technical expertise and large budgets. The authors first delineate the technical distinction of GAI: large‑scale language models (LLMs) and multimodal generators create new text, images, video, or code from massive, largely unlabeled training data, using simple prompts rather than task‑specific training. This capability democratizes high‑end AI functions that were previously the preserve of large corporations.

The authors then enumerate six “promises” that GAI can deliver to SMEs: (1) productivity gains through automation of routine outputs such as reports, contracts, and accounting entries; (2) rapid access to up‑to‑date human knowledge via real‑time summarisation of market, legal, or technical information; (3) a boost to creativity by enabling fast prototyping of marketing copy, visual assets, and product concepts; (4) differentiation of products and services through personalised recommendations and design; (5) acceleration of product development via AI‑assisted simulation, scenario analysis and hypothesis testing; and (6) enhanced decision‑making through large‑scale data analytics and predictive modelling. Because most GAI tools are offered as low‑cost SaaS or API services, the financial barrier for SMEs is modest compared with traditional AI projects.

Conversely, the paper outlines six categories of challenges. Adoption hurdles stem from limited budgets, scarce data‑science talent, and inadequate IT infrastructure. Accuracy concerns arise from the well‑documented “hallucination” problem of LLMs, which can produce confident but false statements—particularly risky in legal, financial, or medical contexts. Ethical issues include bias, privacy infringement, and copyright violations. Reputational and legal risks emerge when erroneous AI output harms customers or breaches regulations. Heavy reliance on third‑party providers raises supply‑chain and data‑security vulnerabilities. Finally, the cumulative financial burden of subscriptions, custom‑model fine‑tuning, and staff training can strain SME cash‑flows.

To integrate these opportunities and risks, the authors introduce a “sailing metaphor” that frames the SME as a ship navigating turbulent, uncharted waters. Four nautical elements constitute the strategic framework:

1. Crew (Employee Competency) – Employees must possess three core capabilities: a learning orientation (ability to self‑direct and unlearn old habits), technology curiosity (willingness to experiment with prompts and tools), and adaptability (flexibility to adjust workflows as AI outputs evolve). The paper recommends internal learning platforms, mentorship, and a culture of rapid prototyping.

2. Captain (Leadership & Values) – Senior leaders must articulate clear GAI objectives, embed ethical and legal guardrails, and champion an AI‑friendly value system that encourages responsible experimentation.

3. Chart (Organizational Culture & Strategy) – A culture that tolerates failure, promotes data transparency, and emphasizes cross‑functional collaboration is essential for sustained AI adoption.

4. Route (Collaboration & External Partnerships) – SMEs should build API‑based ecosystems, engage with specialist vendors, and participate in joint research or consortiums to mitigate the risks of single‑vendor dependence.

The paper provides a four‑stage implementation roadmap:

• Diagnosis – Map existing processes, identify high‑impact pilot use‑cases, and assess readiness.
• Pilot Execution – Deploy a minimum viable product (MVP) of the chosen GAI application, iterate quickly, and collect feedback.
• Scale‑Up – Roll successful pilots across the organization, formalise training and certification programs, and invest in up‑skilling.
• Sustain & Govern – Establish KPIs, continuous monitoring, and an ethics‑compliance board to oversee long‑term performance and risk mitigation.

By situating the discussion within the sailing metaphor, the authors not only offer an intuitive visualisation of complex strategic interdependencies but also extend AI‑management theory to explicitly account for SME‑specific constraints such as flat hierarchies, multi‑role employees, and limited resource buffers. The paper thus makes a dual contribution: (1) a scholarly synthesis of GAI’s potential and pitfalls for SMEs, and (2) a practical, step‑by‑step guide that SMEs can adopt to harness generative AI responsibly while safeguarding competitiveness and societal trust.