Recently, Microsoft analysis shows that AI technology has been successfully applied in more than 700 real business scenarioses around the world. These cases span many industries such as manufacturing, retail, healthcare, finance, etc., and cover large multinational companies and small and medium-sized enterprises, including Fortune 500 companies. Some studies have shown that for every £1 invested in generative AI, companies can earn an average return of £3.7. Such an attractive ROI is undoubtedly accelerating the embrace of AI in all walks of life.

Based on 700 real-world cases, this paper will classify and analyse the industry distribution, application types and technology trends of AI applications, and focus on how the practice of AI in the field of enterprise management (such as ERP, finance, supply chain, etc.) inspires and serves Chinese enterprises.

1

Industry distribution: AI applications are blooming everywhere, with finance and manufacturing leading the way.Judging from the more than 700 AI cases collected, AI empowerment has penetrated almost all major industries. Among them, the financial industry and manufacturing industry are the most active areas of AI application, accounting for about 21% and 19% of all cases. This is followed by technology/communications, services and other industries, which shows that AI is no longer a “patent” of one or two industries, but is becoming a universal enabling technology across industries. If AI is compared to new infrastructure – just like electricity was to the industrial revolution in the past, AI is now becoming a “new public resource” for enterprise digital transformation.

2

Application Types: From customer service to prediction, AI plays a diverse role Combing through 700 cases, it can be found that enterprise application AI is mainly concentrated in several typical scenarioses, each of which corresponds to different focuses on improving business value.

(1) Customer Service and Marketing Automation: A large number of businesses use AI chatbots and virtual assistants to optimise customer interactions. For example, Indonesia’s BRI Bank has integrated AI technology into its customer chatbot, supporting multiple local languages and significantly improving customer satisfaction. Retail e-commerce platform 17Life uses AI to automatically generate and classify product tags to accurately understand consumer search intent, successfully improving the efficiency of personalised recommendations. In the field of marketing, Unilever has developed a marketing assistant that automatically aggregates market data and generates creative proposals, greatly speeding up the planning and execution of advertising campaigns. By providing 24×7 hours of quick response and personalised content, these AI applications not only reduce the burden on manual customer service and marketing teams, but also significantly improve customer experience and conversion rates.

(2) Employee efficiency and collaboration improvement: This is one of the most widely used types of AI, and many organisations equip employees with AI “partners” to help alleviate repetitive daily tasks and allow employees to focus on higher-value tasks. For example, after the Bank of Queensland in Australia piloted AI Copilot, 70% of employees saved 2.5 to 5 hours per week; the Somerset County Government in the United Kingdom deployed Copilot to save employees 10 hours per month, and 87% of users reported that the work was easier. KPMG, a world-renowned accounting firm, has developed an AI assistant for employee onboarding that automatically provides templates and reference materials for new employees, reducing the preparation time of training materials by 20%. For example, after embedding GitHub Copilot into the development process, programmers have significantly accelerated coding and problem-solving, and collaboration efficiency has also been significantly improved. It can be said that AI assistants are becoming the right-hand men of employees, from writing documents and summarising meeting minutes to collecting information.

(3) Business process optimisation and automation: Many enterprises embed AI into their core business processes, driving a qualitative leap in operational efficiency. The most typical aspect of this is enterprise management and operation processes, covering finance, supply chain, production, internal support, etc. For example, Animal Supply Company, an animal supply distribution company, used an AI document intelligence platform to revolutionise manual invoice processing, saving $500,000 annually and freeing up 50% of invoice specialists’ time for exception handling and supplier relationship building. In manufacturing, Bridgestone introduces AI into factory maintenance, combining sensor data and large models to enable predictive maintenance and natural language queries, improving production continuity and reducing downtime. Another example is India’s ICICI Lombard insurance company Copilot assistant for claims adjusters, automatically extracting key points from massive claim documents, reducing the time to process a claim by more than half. It can be seen that AI is penetrating into business processes, making enterprise operations more efficient, accurate and agile by automating repetitive tasks, intelligently parsing data, and optimising decision-making processes.

（4）Intelligent forecasting and decision support: Enterprises are leveraging AI’s powerful analysis and forecasting capabilities to make more informed business decisions. In supply chain management, AI can perform demand forecasting and inventory optimisation based on historical data and real-time information. For instance, European retailer SPAR has developed an AI-powered demand forecasting system with an inventory forecasting accuracy rate of up to 90%. ACWA Power, a company in the energy sector, combines AI and IoT to realise real-time monitoring and predictive maintenance of power generation equipment, reducing maintenance costs and improving operational safety. Financial investment institutions have also introduced AI-assisted risk control decision-making, such as UBS, a Swiss bank, which uses AI to create a legal AI assistant to help employees find compliance information faster and improve the efficiency and accuracy of risk assessment. In summary, the use of AI for intelligent prediction enables enterprises to more proactively optimise supply chains, improve equipment reliability, gain insight into market trends, and prevent risks, thereby seizing opportunities in the fierce competition.

（5） Innovative R&D and product and service upgrades: AI is becoming the engine of enterprise innovation, accelerating product development and new business incubation. Some pharmaceutical and scientific research institutions have used generative AI to significantly shorten the R&D cycle, such as Japan’s Daiichi Sankyo, which built an internal large model DS-GAI in just one month and promoted it throughout the company, improving the efficiency and accuracy of R&D personnel. The gaming and media industry is also using AI to unleash creativity: Square Enix has developed a Slack chatbot to provide instant technical Q&A support for game developers, accelerating game production; Four Agency, a British media company, has introduced AI to spark creative ideas, accelerate data analysis and report generation, allowing employees to spend more time on market expansion. It is foreseeable that AI will continue to drive innovation in business models and product forms: from design prototypes and code development to content creation and personalised services, all of which are faster and more efficient because of AI. This provides businesses with unprecedented acceleration of innovation.

3

Technology trends:

Led by large models, integrating multiple AI technologiesThrough case analysis, it is not difficult to find that the technical direction of today’s enterprise-level AI applications shows distinct trend characteristics. Many cases reflect the trend of multi-technology integration. Enterprises often combine large models with RPA (robotic process automation), knowledge graph, computer vision (CV), speech recognition, etc. to create comprehensive intelligent solutions. For example, the field sales assistant developed by HEINEKEN integrates speech recognition and computer vision: sales representatives use AI voice to serve multilingual voice input, AI automatically records the outlet situation and triggers back-end processes, and uses CV technology to intelligently parse documents, improving the data collection efficiency of front-line sales.

4

Case Insights:

AI empowers a new paradigm in enterprise managementIt is worth noting that more than 700 cases show that AI is fully penetrating finance, supply chain, human resources and other management links, bringing revolutionary changes to the traditional enterprise operation model.In China, AI is also setting off a “digital intelligence revolution” in the field of enterprise management, especially in the central links of enterprise management such as finance and supply chain.When it comes to smart finance, AI has a lot to say. The finance department has long undertaken a lot of tedious and repetitive work, such as accounting review, invoice processing, report preparation, etc. The introduction of AI is changing this, such as the Colombian BDO company, which halves the workload of payroll and financial processes through virtual assistants, which is a typical example of financial AI applications. In the field of invoices and bills, Animal Supply in the United States uses AI to save $500,000 in costs per year, improving the automation and accuracy of the financial shared service centre. The enlightenment of these practices is that financial work can be qualitatively changed through “AI financial assistants”.

AI can quickly read bills and vouchers, check for errors, and even provide financial insights based on historical data. For example, Kingdee’s intelligent financial solution aims to allow every employee to have an AI financial assistant, providing assistance from travel agents, intelligent document review, financial data insight, analysis to report writing through the perception, memory, and decision-making capabilities of large models. For financial personnel, Kingdee’s AI assistant can be embedded in financial process nodes to provide real-time guidance, and help complete tasks such as attachment information refinement, comparison, and review suggestions through sidebar tools to achieve process reshaping. For management, AI assistants can also instantly summarise and analyse key financial indicators to help decision-makers fully grasp the business situation. It is conceivable that in the near future, “AI accounting” and “AI auditing” will become standard, and financial personnel will be freed from a lot of mechanical work and devote their energy to strategic financial planning and business cooperation. This will undoubtedly greatly improve the strategic value of financial management.

In the field of smart supply chain and ERP optimisation, AI also makes great progress. Supply chain management involves procurement, inventory, production planning, logistics, and other links, and AI can optimise these complex processes through prediction and automation. For example, the “TAMI” system developed by manufacturing giant Textron Aviation allows front-line engineers to search for maintenance instructions from 60,000 pages of technical documents in a conversational way, reducing troubleshooting time from 20 minutes to less than 2 minutes – which is actually a model of combining ERP/knowledge base with large models. Icertis provides AI-driven contract review, risk assessment, and hidden clause mining for 30% of Fortune 100 companies, covering multiple scenarioses such as contract lifecycle management (CLM), agreement data insights, legal due diligence, internal financial compliance review, and semantic search for union agreements.

For Chinese manufacturing and supply chain enterprises, this means that by introducing AI, a comprehensive upgrade from demand forecasting, procurement contracts, production scheduling to inventory management can be achieved, reducing supply chain management costs and improving response speed to market changes. For example, Kingdee Cloud Xinghan is based on large model technology to help enterprises achieve intelligent contract review, automatic supplier recommendation, intelligent sourcing, intelligent replenishment, etc.

It can be seen that the application of AI in the field of enterprise management is triggering a change in the management paradigm – from “human rule” to “human-machine collaboration”. For Chinese companies, this is both a challenge and a huge opportunity. As China’s top enterprise management AI company, Kingdee has provided AI empowerment solutions in line with national conditions for many local enterprises. For example, Kingdee Cloud Xinghan provides intelligent assistants for finance, human resources, procurement and other fields through the built-in Sky Agent platform, and deeply embeds AI into business processes. Recently, Kingdee has released five major agents, including Golden Key Financial Report (Financial Report Analysis Agent), ChatBI (Enterprise Question Intelligence Agent), Recruitment Agent, Travel Agent and Enterprise Knowledge Agent, covering multiple scenarioses where AI is most widely used on the enterprise side.

More than 700 real-world cases around the world vividly verify that AI is no longer a cutting-edge technological experiment, but has become a practical business productivity tool. From enhancing employee efficiency and optimising customer experience to reshaping business processes and giving rise to innovative models, the value that AI brings to various industries and functions is obvious to all. For Chinese companies, these global practices also provide valuable inspiration and confidence.