Table of contents

• Key Takeaways
• GenAI adoption: A spotlight on major surveys
• GenAI adoption: Emerging industries
• Appendices

Key takeaways

The state of GenAI in the enterprise

• Strong potential for value addition tempered by long timelines for implementation: Enterprise GenAI adoption can unlock an annual equivalent of USD 2.6 trillion–4.4 trillion in value across the global economy, but the timeline for realizing these benefits is drawn out, with ~60% of US executives and ~68% of global business leaders indicating their organizations were still at least 1–3 years away from implementing their first GenAI solution and witnessing the benefits of transformation. 

• Adoption led by digitally inclined, knowledge-based industry verticals: Global GenAI adoption is led by technology, media, and telecom (TMT), followed by the financial services sector. Middle-of-the-pack adopters include consumer goods/retail, aerospace and defense, pharma and healthcare, and energy and utilities.

• Highest value to be derived from four business functions: An estimated ~75% of the value to be derived from GenAI use cases is expected to stem from four areas: 1) customer operations, 2) marketing and sales, 3) software engineering, and 4) R&D. 

• Industry alliances can accelerate enterprise adoption: Developer-led collaborations (between GenAI developers and tech partners) and consultancy-driven partnerships (between consulting firms and GenAI developers) can play an important role in enhancing the general accessibility of GenAI solutions as well as in supporting and scaling enterprise adoption, respectively. 

GenAI adoption in emerging industries

• Highest incidence of GenAI activity in healthcare and pharmaceuticals: Emerging industries engaged in providing and improving healthcare services, such as Telehealth, Hospital Interoperability, and Hospital Management, showed the strongest propensity for adoption, with a range of GenAI-focused activities focused on either developing the technology in-house or leveraging third-party GenAI. 

• High activity also observed in financial services: In line with broad global patterns around knowledge-based industries, emerging financial services industries like InsurTech, FinTech Infrastructure, and to a lesser extent, Neobanks, also showed a high incidence of GenAI activity. 

• Strongest potential for value creation through three business functions: Core business operations, research and product development, and customer experience witnessed the highest levels of GenAI activity across the emerging industries. By extension, companies looking to generate value through GenAI adoption need to prioritize relevant, industry-specific use cases across these functions. In healthcare, for example, enabling GenAI for core business operations may involve use cases like patient care administration or supporting diagnostics and treatment. Comparatively, in financial services, it may include underwriting or fraud detection.

• Further value addition may be driven by industry-agnostic GenAI: Our research suggests that companies in emerging industries have also leveraged GenAI to develop solutions for general business functions. This includes industries like Marketing Automation (for customer analysis and content creation), Next-gen Cybersecurity (to develop security assistants, automate coding, and bridge data silos), and Supply Chain Tech (for item description generation, supplier recommendations, and negotiation summaries). The industry-agnostic nature of these solutions has likely supported the current high enterprise adoption in marketing and sales as well as software engineering. It also indicates emerging industries could accelerate GenAI enablement beyond the three fundamental business functions.

GenAI adoption: A spotlight on major surveys

The state of GenAI in the enterprise

• In theory, enterprise GenAI adoption presents significant quantifiable economic benefits. In 2023, McKinsey estimated that GenAI organizational use cases could unlock an annual equivalent of USD 2.6 trillion–4.4 trillion in value across the global economy, with GenAI-enabled enhancements to worker productivity further driving this to USD 6.1 trillion–7.9 trillion annually.

• In reality, however, organizations are still in the early stages of scaling their GenAI investments relative to other software categories. According to Menlo Ventures’ 2023 enterprise survey (450+ executives across the US and Europe), enterprise investment in GenAI stood at a comparatively humble USD 2.5 billion in 2023, relative to sizable enterprise budgets for traditional AI (USD 70 billion) and cloud software (USD 400 billion). Around 80% of the executives surveyed reported a preference to buy rather than build GenAI solutions and purchased third-party software. In 2023, an estimated 40% of the overall USD 2.5 billion enterprise investment in GenAI was channeled toward incumbents like Microsoft and tools like GitHub Copilot.

• While the precise value created by GenAI remains to be seen, there is widespread organizational interest in the technology. In April 2023, Capgemini’s research indicated that GenAI was on the boardroom agenda at 96% of ~1,000 global organizations surveyed, with ~60% of executives reporting that their leadership strongly advocated for the technology and only ~39% taking a “wait-and-see” approach to adoption. Meanwhile, around one-third of survey respondents in McKinsey’s Global Survey of April 2023 (1,684 enterprises) indicated that their organizations were regularly using GenAI in at least one business function. 

• The timeline for GenAI implementation among US organizations, however, remains drawn out. While ~65% of respondents in KPMG’s March 2023 survey (~225 US executives) believed GenAI would strongly impact their organization over the next 3–5 years, ~60% also thought they were still a year or two away from implementing their first GenAI solution. 

• The same extended timeline is also evident at the global level. Around 38% of respondents in EY’s Reimagining Industry Futures Study in November 2023 (1,405 enterprises worldwide) favored a measured, incremental approach to GenAI adoption, despite the technology’s popularity. According to Deloitte’s December 2023 survey (2,800+ global business leaders), ~68% expected a timeline of more than a year for GenAI to truly transform their organizations.

• Notably, organizations with extensive AI experience tend to have more conservative timelines for realizing the transformative impacts of GenAI. In October 2023, EY’s CEO survey (1,200 CEOs globally) indicated that ~66% of organizations that have successfully implemented at least one AI initiative expect the technology to transform their business and operational models within two years. However, organizations with more extensive AI experience (5+ AI-related initiatives) had a more cautious timeline of 3–5 years.

GenAI adoption across industry verticals

• TMT and financial services tend to lead in terms of global GenAI adoption. According to McKinsey’s Global Survey on AI, TMT stood ahead of the pack in GenAI usage for work (14%), followed by financial services (8%), business, legal, and professional services (7%), and consumer goods/retail (7%). 

• In the US, TMT again features as a prominent GenAI adopter. KPMG’s research shows that 60% of US respondents in TMT considered it a high or extremely high priority to research GenAI applications over the next 3–6 months (the highest of all industries). Meanwhile, respondents from TMT and financial services were also the most likely to report that tools like ChatGPT have had a large impact on their digital and innovation strategies.

• Nonetheless, global GenAI adoption is not limited to TMT and financial services. Several other industry verticals also appear to be gearing up for adoption. Capgemini’s findings mirror the broad trends picked up by McKinsey’s research, with high tech leading the way (74%) in terms of dedicating resources to GenAI integration in product development. However, several other industries, including retail (62%), aerospace and defense (52%), pharma and healthcare (42%), financial services (42%), energy and utilities (39%), and telecom (36%), also appear to be preparing for GenAI adoption.

• While all industries are likely to experience a degree of disruptive change from GenAI, the ones relying heavily on knowledge work are expected to be the most impacted. Knowledge-based industries like banking, technology, pharmaceuticals and medical products, and education are likely to experience disruptive impacts. By contrast, manufacturing-based industries may experience less disruptive effects.

• Language models, in particular, can be deployed to great effect for both external and internal stakeholders in knowledge-based industries like financial services. In November 2023, for instance, NatWest was one of the first banks in the UK to deploy a GenAI-enabled virtual assistant in partnership with IBM. The new solution, Cora+, was able to access textual content from multiple secure sources, letting customers compare solutions across the bank’s product suite and obtain information across NatWest Group websites. Meanwhile, AXA released AXA Secure GPT in partnership with Microsoft Azure OpenAI in July 2023. By April 2024, the platform had been rolled out globally to ~140,000 AXA employees, enabling a range of text generation and translation tasks including the creation of job descriptions, communications drafts, and code generation in several programming languages.

GenAI adoption across business functions

• Marketing and sales tends to feature prominently in terms of GenAI adoption by function. According to McKinsey’s Global Survey on AI, the most commonly reported uses of GenAI tools were in marketing and sales (14%), product and service development (13%), and service operations (10%). 
• The development of drafts and summaries of text documents was a common use case across these functions. Other use cases included personalized marketing, the development of new product designs, trend identification, the use of chatbots, and forecasting service trends or anomalies.
• Across industries, several other business functions also appear to be gearing up for GenAI adoption globally. Capgemini’s findings reveal high GenAI adoption in IT (54%) and echo McKinsey’s results on the high adoption of GenAI by sales and customer service (47%) and marketing and communications (46%). Functions like manufacturing (30%), product design (28%), operations (24%), and risk (21%) are also seen to be in play.

• Even so, most enterprises are likely to derive the highest value from leveraging GenAI in select business functions. McKinsey estimates that as much as ~75% of the value to be derived from GenAI use cases is likely to stem from four areas: 1) customer operations, 2) marketing and sales, 3) software engineering, and 4) R&D.

The GenAI adoption dilemma

• Several push-pull forces appear to be impacting enterprise GenAI adoption across industries. According to Gartner surveys, this includes
  1. Board and CEO expectations: ~71% of business executives implementing GenAI reported that the push to implement was mostly top-down. 
  2. Customer expectations: ~65% of consumers reported being comfortable with or neutral about the use of GenAI in marketing, while 56% reported the same for the use of GenAI in customer service. 
  3. Employee expectations: ~50% of employees want to be able to delegate routine administrative and physical tasks to AI fully.
  4. Investor expectations: With investors rewarding organizations for growth and productivity, organizations are increasingly driven to pursue the efficiency gains enabled by GenAI. For instance, Gartner’s November 2023 poll of 821 business executives implementing GenAI revealed projections of 15.7% in cost savings and 24.69% in team productivity improvements over the next 12–18 months.
  5. Regulatory expectations: On the flip side, ~68% of senior business leaders are concerned about GenAI surpassing organizational risk mitigation capabilities.

• Potentially gaining a competitive edge may drive enterprises to GenAI adoption, but uncertainties around the technology continue to present barriers to implementation. According to EY’s CEO survey, 62% of business leaders acknowledge the urgency of acting on GenAI to gain a strategic edge over competitors. However, ~61% have concerns about the formulation and execution of an AI strategy.

• Key barriers include cost, the lack of a clear business case, cybersecurity, data privacy, and the absence of appropriate risk mitigation and AI governance strategies. According to KPMG’s 2023 survey of US executives, respondents from businesses with revenues of USD 1 billion or more cited cost and the lack of a clear business case as their biggest barriers to implementing GenAI. Respondents also indicated concerns about cybersecurity (81%) and data privacy (78%). While 72% of executives believed GenAI had the potential to positively impact stakeholder trust, nearly 45% remained concerned that the technology could also have negative effects in the absence of appropriate risk management measures. As of 2023, only 6% of organizations surveyed had set up a dedicated risk mitigation team and only 5% had a mature responsible AI governance program in place.

• There also tend to be challenges around the practical aspects of adopting and scaling GenAI. According to McKinsey, 90% of GenAI pilots fail to reach full production due to complexities associated with adopting and scaling the technology. This includes dealing with unstructured data, developing advanced algorithms, building a suitable IT architecture, driving capacity building and domain expertise, and ensuring effective change management.

Notable alliances driving GenAI adoption 

• There appear to be two broad types of alliances driving GenAI adoption across industries: 1) developer-led collaborations and 2) consultancy-driven partnerships. Most of these alliances apply across industries, with a few key verticals being prioritized including financial services, tech, retail and consumer packaged goods, and manufacturing. The table below provides some examples of these alliances.

• Developer-led collaborations usually involve strategic technology partnerships between major GenAI developers and other collaborators working together to expand cross-industry adoption. A classic example is the partnership between Microsoft and OpenAI to offer enterprises access to GenAI via the Azure OpenAI service. A more recent alliance is Microsoft’s partnership with digital transformation company Cognizant in April 2024 to make its GenAI and Copilots available to millions of enterprise customers across industries, contributing an estimated USD 1 trillion from AI into US GDP over the next 10 years. 

• Consultancy-driven partnerships also include GenAI developers but involve consulting firms in the integral role of supporting and scaling enterprise adoption. McKinsey, for example, has set up an enterprise GenAI ecosystem with ~18 GenAI technology providers including Anthropic and Cohere. The ecosystem functions as a central hub for enterprise clients to access expertise and solutions from these ~18 providers across all parts of the tech stack, ranging from cloud infrastructure to LLMs. 

GenAI adoption: Emerging industries

GenAI adoption across industry verticals

• Enterprises across some of SPEEDA Edge’s emerging industries prioritize GenAI differently from those featured in third-party, cross-industry surveys. At least some of these observed variations may be chalked up to the strong inherent focus on next-gen tech among emerging industry players, from digital-native startups to tech-savvy incumbents. Comparatively, the third-party surveys feature a broad sample of enterprises, including relatively more traditional players.

• Healthcare and pharmaceuticals lead the pack in GenAI adoption among emerging industries. Enterprises in this industry vertical are especially inclined to embed GenAI, whether this involves developing the technology in-house or leveraging third-party GenAI. In AI drug discovery, for instance, in silico drug discovery startup AQEMIA has leveraged its in-house GenAI-enabled deep physics platform in partnership with Sanofi to accelerate the R&D of small molecule drug candidates. Meanwhile, telehealth provider Pager Health has developed three customer-focused applications using Google Cloud’s GenAI, summarizing member interactions (“Chat Summation”), answering common health and benefits queries while reducing the burden on call centers (“FAQ Bot”), and analyzing member sentiments via chat to boost care service quality (“Sentiment Analysis”).

• The financial services sector is another strong GenAI adopter among emerging industries. For instance, US neobank Dave partnered with enterprise GenAI provider Aisera to release DaveGPT, a GenAI assistant that offers self-service customer inquiry resolution for Dave’s 9.9 million members. Similarly, no-code insurance platform Instanda partnered with conversational GenAI platform Floatbot.AI to further engage policyholders through multimodal conversational AI agents. 

GenAI adoption across industry verticals and business functions 

• Core business operations, research and product development, and customer experience are the business functions seeing the highest GenAI adoption. Emerging industries primarily tend to leverage GenAI to optimize core business operations (e.g., using GenAI in financial services to streamline loan processing or insurance underwriting), develop new and monetizable products (e.g., using GenAI to discover new plant-based meat formulations in food and agriculture), and boost customer experiences (e.g., using GenAI in financial services to incorporate chatbots to engage policyholders). 

Deep dive: Major GenAI adopters 

Healthcare and pharmaceuticals

• GenAI has impacted the core business operations of industries engaged in providing and improving healthcare services, such as Telehealth, Hospital Interoperability, and Hospital Management. Common use cases include: 
  1. Patient care administration (as seen with Teladoc, Amwell, and Oracle Cerner): Patient record-keeping, managing and searching clinical documents, and prescription validation. 
  2. Data aggregation and insights (such as developments from Xsolis, Redox, and EPIC): Creating patient profiles and identifying the most suitable patients for care. 
  3. Supporting diagnostics and treatments (as done by Amdor Health, BrightInsight, and Glooco acquiring xbird): Remote patient monitoring, identifying medication and treatment-related issues, identifying health risks, and assisting caregivers.

• GenAI for research and product development was more common among industries engaged in pharmaceutical development, such as AI Drug Discovery, Precision Medicine, and Clinical Trial Tech. AI Drug Discovery companies including Vant AI, Absci, and BioPhy commonly use GenAI to design, understand, and predict molecular structures for drug development; develop therapeutic antibodies (through biological research and developing novel protein-based therapeutics); research vaccines; and offer scientific and clinical guidance to support regulatory compliance, quality assurance, and other drug development functions. Precision Medicine companies adopted GenAI for genomic data analysis (Dante Genomics), pathology diagnostic data research (Quest Diagnostics), and to gain insights on patient and molecular information (Tempus). Finally, Clinical Trial Tech leveraged GenAI to analyze data to improve patient-trial matching (Triomics) and for data quality checks (Saama Technologies).

• Healthcare service industries like Beauty Tech, Telehealth, Longevity Tech, and Age Tech have also used GenAI to improve customer experiences. This has primarily been achieved through digital assistants and chatbots that onboard customers, respond to their queries, and offer personalized interactions. In addition to customer assistants, Beauty Tech companies have notably used GenAI for virtual product try-ons, mainly for skincare products and hairstyles.

Financial services

• GenAI has primarily impacted the core business operations of emerging financial services industries like InsurTech (Personal Lines, Commercial Lines, and Infrastructure), FinTech Infrastructure, and to a lesser extent, Neobanks.
  1. InsurTech has seen adoption in underwriting (for instance, by Swiss Re, Dais Technology, and Zesty AI for risk assessment, personalized quote development, and to increase processing efficiency and quality); claims processing (as used by Clearcover, Sprout.ai for streamlining statement collection and claims assessment), including data extraction and document processing that supports both activities; and insurance fraud detection and prevention. 
  2. Meanwhile, FinTech Infrastructure has seen the highest adoption in fraud detection and prevention (as used by Airwallex and Socure, including for identity verification and KYC assessments) and a few use cases in improving payment processing and loan underwriting.
  3. Within Neobanks, GenAI has mostly been adopted for compliance checks (Grasshopper) and employee support (Lunar).   
• Emerging financial services industries have also adopted GenAI to improve customer experiences. Neobanks are among the foremost adopters of GenAI-enabled virtual assistants and chatbots. Some assistants have been developed to autonomously deal with customer inquiries and replace search functions, offering personalized solutions based on customer profile and eligibility, while others have been developed to support customer representatives. InsurTech companies have also used GenAI to develop solutions for insurance agents to better engage with their customers.  
• Meanwhile, Business Expense Management stands out as a notable GenAI adopter across functions like finance and accounting as well as supply chain management. Common use cases in finance and accounting include 1) automating expense workflows (from invoice capture to approvals and fraud detection), 2) information audits (such as purchase order, ledger, and invoice matching), 3) data extraction and expense report creation, 4) expense insights, and 5) expense policy enforcement. Supply chain management use cases on the other hand primarily involve enhancing supplier relationships through improved communication and insights, vendor management and negotiations, and vendor contract reviews.   

Manufacturing and retail

• GenAI has been embedded in advanced manufacturing technologies, primarily to improve core manufacturing operations. 
  1. Smart Factory has seen GenAI adoption in several use cases, including 1) improving industrial robotics (coding and programming, simulations, supporting complex tasks, and enhancing object identification and path recommendations), 2) industrial automation system design and development (including code generation through natural languages), 3) preventive maintenance, and 4) generating industrial insights (operational insights, asset performance optimization, report generation).
  2. The Digital Twin industry has also leveraged GenAI to deliver better operational insights and preventive maintenance, minimizing plant downtime and improving plant management and control. 

• Meanwhile, tech providers in the Additive Manufacturing and Digital Twin industries have used GenAI to improve product design and development for its users. 
  1. Additive manufacturing companies such as Autodesk, ToffeeAM, and Hexagon have enabled users to design products and industrial parts based on users’ inputs of design goals and constraints (also capable of developing multiple design alternatives based on these parameters). These can also be designed to improve performance (such as thermo fluid and structural performance).  
  2. GenAI-powered digital twins were used by NVIDIA to design, build, and validate products, manufacturing processes, and factories virtually before creating them in reality. 

Transportation and logistics

• Auto Tech remains the largest GenAI user in the transport and logistics space, driven by a strong focus on boosting customer experience. For the most part, this has been achieved through GenAI-powered voice assistants (as developed by Volkswagen, Continental, and General Motors, among others) that provide advanced general and navigational support through context-aware conversations and by accessing specific information on the vehicle and its functions. Meanwhile, GenAI-powered chatbots have been used by Truck Industry Tech solutions providers such as Geotab to offer personalized insights for fleet management and by Toyota in the EV Economyto offer advanced product visualization tools (picturing a vehicle in different environments).   

• Research and product development has also seen relatively high adoption. Industries like Auto Tech and Truck Tech have leveraged GenAI-based solutions to develop better autonomous driving and driver assistance solutions, analyze and manage cyber data and alerts, and develop predictive analytics software for freight logistics. Meanwhile, Toyota and LG Energy Solution in the EV Economy have adopted GenAI to design vehicles that conform to engineering and manufacturing constraints and to research the latest tech trends from third-party sources.

Quick look: Slow GenAI adopters 

• Food and agriculture: We have seen multiple companies featured in the Plant-based Meat and Plant-based Dairy & Egg industries, such as NotCo, Shiru, and The Live Green Co, leveraging GenAI for product formulation and development. GenAI has also been harnessed to assist the core operations of Smart Farming through crop management analytics; grass height measurement; and weather, soil, and crop cycle management.

• Energy and climate change: The space has seen GenAI adoption in Carbon Management Software (for carbon accounting and management) and Hydrogen Economy (for product design and performance optimization).  

Other notable GenAI adopters: Focus on industry-agnostic tech

Emerging industries with industry-agnostic, GenAI-enabled solutions

• Marketing Automation and Sales Engagement Platforms typically leveraged GenAI for several use cases including
  1. Customer analysis (such as Twilio, Baidu, and Lytics) for audience targeting and segmentation. 
  2. Communication content and workflow automation(such as Slack, Zoho SugarCRM, Adobe), for developing images, titles, and descriptions; customized communication content; and delivering SEO-based content.
  3. Generating sales insights (as done by Zoominfo, and Customers.ai) for prospecting, performance analysis, and analyzing leading indicators for revenue team success. 
  4. Campaign design and management (as done by Stagwell, MoEngage, and Mailchimp).
  5. Customer sentiment analysis (such as Estee Lauder partnering with Google Cloud to develop solutions to better understand customer sentiment).

• Next-gen Cybersecurity has leveraged GenAI primarily to develop assistants that help security practitioners, developers, and compliance teams, among others, to improve their security practices. This includes defining security policies, identifying and investigating threats and vulnerabilities, offering mitigation advice, and monitoring. GenAI also helps automate coding to develop models, bridge data silos, enhance attack path simulations, as well as predict and respond to data breaches.  
• Meanwhile, GenAI has been used by Workflow Automation Platforms for finance and accounting automation as well. This includes automating accounting workflows (accounting, revenue recognition, lease accounting, and close management) as well as variance analysis and reconciliation automation. 
• Extended Reality (XR) solutions have seen relatively high GenAI adoption in the research and product development for the media industry, healthcare, and retail. In the media industry (gaming, social media, and entertainment), GenAI-enabled XR helps users create filters, games, and characters using text prompts and images. In healthcare, GenAI-enabled XR has been used to streamline research, and in retail, for virtual try-ons and creating virtual renditions of products. The potential to use GenAI to create alternative reality (AR) content and images may impact more industries in the future. Additionally, GenAI-based XR solutions have also impacted core business operations of the industrial sector, as it relates to training and development and operational support for front-line and field-service workers. More general applications also include incorporating safe mental health breaks into the users’ VR-based programs.  

• Data Infrastructure and Analytics providers have harnessed GenAI to enable users to better interact with data through natural languages, define data quality expectations, and develop actionable business insights. 

• Supply Chain Tech leveraged GenAI to offer dashboards and user insights and for procurement assistance and process automation (including item description generation, supplier recommendations, and negotiation summaries). GenAI has also been used for contractor compliance checks, forecasting stockouts, and identifying supply chain risks through queries.

• Digital Privacy developed GenAI-enabled solutions to mitigate vulnerabilities when enterprises share sensitive information. This also includes protecting sensitive data when using GenAI and large language models (LLMs). 

Appendices

Appendix 1

• Focus: This study tracked GenAI adoption for a subset of 61+ emerging industries on the SPEEDA Edge platform, prioritizing notable target verticals including 1) food and agri, 2) financial services , 3) manufacturing and retail, 4) energy and climate change, 5) transportation and logistics, 6) health and pharma.  

• Definition: GenAI adoption was defined in terms of the number of GenAI-focused activities undertaken by incumbents and startups across the emerging industries. 

• Activities tracked: 1) partnerships (to develop GenAI-based solutions), 2) in-house product development (included the addition of GenAI updates to existing products, the introduction of wholly new GenAI-enabled products, and the release of plans to launch GenAI-based products), and 3) M&A (focused on improving GenAI solutions). Activities associated with traditional AI were excluded from consideration.

• Time period: January 2023–April 2024

• Use cases: The study considered GenAI-focused activities targeted at external use cases (catering to B2B or B2C customers) and internal use cases (focused on improving business processes such as training or operational insights).

Appendix 2

Appendix 3