The Interplay of AI Adoption, IoT Edge, and Adaptive Resilience to Explain Digital Innovation: Evidence from German Family-Owned SMEs

Abstract

This study aims to discover how artificial intelligence adoption in notion (AI) plays a role in digital innovation using the theoretical foundation of diffusion of innovations and effectuation theories. The current research also investigates the moderating role of other edge Internet of Things (IoT) and the mediating role of adaptive resilience. The data collection is performed using a survey conducted among employees of family-owned SMEs. The findings reveal that AI forecasts digital innovation through adaptive resilience. The results also confirm the moderating role of threat to IoT edge and the mediating role of adaptive resilience, but moderated mediating is not supported. We conclude that family-owned SMEs intend to adopt AI, but SMEs face challenges using IoT edge. This study has implications for family firms specifically and technology adopters in general.

1. Introduction

Digital innovation (DI) has become increasingly crucial for multinational firms and family-owned small and medium enterprises (SMEs) in Europe to reach global customers due to rapid technological advancement and the proliferation of digital platforms and artificial-intelligence-related tools [1,2]. DI could use limited SME resources to create novel products, services, and business models for family-owned SMEs [3]. In family-owned SMEs, digital innovation can also help address some unique challenges these firms face, such as limited resources and balancing family and business interests [4]. Finally, digital innovation [5] could drive transformation in SMEs by applying digital technologies, such as artificial intelligence adoption [6], free big data from customers on the internet [7], and the Internet of Things [8,9].

Despite the growing importance of DI, more clarity is needed on how family SMEs are promoting DI with limited resources.

Scholars researching AI adoption in family SMEs say that while entrepreneurship is the primary driver of economic development and industry transformation, combining digital technology [8] and entrepreneurship provides new opportunities for family enterprises to innovate digitally [10]. As digitization has several vital implications, family SMEs and researchers should be aware of potential usage, outcomes, and associated opportunities for AI adoption for resilience and survival in the market for extended periods [6]. Compared with other known technologies, the dominance of artificial intelligence (AI) and machine learning outcomes in digital innovation and disruption is deemed substantial [4]. Family SMEs find viable and potential opportunities to employ AI at every market level. Certain variables, however, may impact the relationship between AI adoption [11], adaptive resilience [12], and digital innovation capabilities of family-owned SMEs [3]. Family business research has limited its ability to describe and comprehend AI adoption in digital innovation [10]. As a result, fresh theories on digital entrepreneurship and plans to accept AI are required to add to the knowledge and context needed to advance the study in this discipline [4,5,11,13].

The study contributes in multiple ways. First, we have contributed by integrating theories of two disciplines. For instance, entrepreneurship’s effectuation theory (ET) is a powerful tool for family SMEs seeking to adopt AI technologies, drive digital innovation, and build adaptive resilience [11,12,13]. Meanwhile, DOI Web 2.0 highlights the role of digital technologies such as AI adoption in shaping the innovation diffusion process to demonstrate better family control and informed decision making [5] through the “network effect”. The network effect could help spread digital innovation for family firms and is usually influenced by the strength of a family firm’s social network and feedback from online communities. The second contribution is contextual, as we specifically studied family-owned SMEs in a technologically advanced Western economy [14]. This study is significant for understanding the behavior of family SMEs towards AI adoption. Although the formation of a family firm is considered an entrepreneurial firm by various scholars [3], scholars say that inheriting a firm raises questions about whether family firms are the same or different in accepting AI due to family control changes across generations and the time of establishment when the start-up was established [1]. As family members and their vision towards technology adoption for better control are crucial to company activities at many levels, family enterprises may implement technology differently depending on their business objectives, geography, culture, and better control of the firms’ resources to avoid agency conflict [1,2,10].

Given the above, the study aims to test the threat to IoT edge’s moderating role and adaptive resilience’s mediating role in line with DOI Web 2.0 [15,16,17] and the effectuation theory [13] using the German family SME’s context. The rest of the article provides dedicated sections for a theoretical underpinning, framework development for family-owned SMEs, results, and discussion.

2. Literature Review

This section is dedicated to developing the study’s framework by integrating diffusion of innovations Web 2.0 [15,16] and the effectuation theory [13].

2.1. Theoretical Underpinning

We use multidisciplinary theories to conceptualize our model for family SMEs, i.e., diffusion of innovations Web 2.0 (DOI Web 2.0) [16], initially proposed by Everett Mitchell Rogers [16], and the effectuation theory of Saras Sarasvathy [13]. DOI Web 2.0 highlights the role of digital technologies such as AI adoption in shaping the innovation diffusion process [4]. Research on family firms’ business has lagged behind other fields such as education and politics in using Web 2.0 tools for innovation, but that is beginning to change. Some critical features of Web 2.0 using DOI include network effects and feedback loops. The “network effect”, which helps spread digital innovation, is influenced by the strength and structure of social networks and online communities and “feedback loops”, which facilitate digital technologies and enable real-time feedback and communication between AI adopters. Therefore, adopting this technical approach can help improve digital innovation, enhance adaptive resilience, and better company control by family owners through modern and cheaper technology adoption as a late adopter compared with the industry [4].

Moreover, we have adopted the theory of effectuation. In the theory of effectuation (2001), Saras Sarasvathy describes a method for making decisions and carrying out actions in entrepreneurial processes, one in which a small firm determines the next, best step by assessing the resources available to achieve the SME’s goal and one in which one continuously balances these goals with one’s resources and actions [13]. The effectuation theory (ET) can be a powerful tool for SMEs seeking to adopt AI technologies, drive digital innovation, and build adaptive resilience [13]. ET can be valuable for family-owned SMEs seeking to adopt AI technologies, drive digital innovation, and create adaptive resilience. For instance, SMEs can take small, affordable steps towards AI adoption and experimentation using free software and websites rather than trying to predict the future [11]. To enhance adaptive resilience, family-owned SMEs can build resilience by using ET to embrace uncertainty and experiment with small, affordable steps by adopting digital technologies and IoT [12]. They can identify potential risks and take proactive measures to mitigate them, such as investing in AI-based free technical solutions or training family firm employees on digital best AI adoption practices [6,8]. They can also build partnerships and networks with other organizations to share knowledge and resources related to adaptive resilience [12]. Similarly, for digital innovation [4], ET can identify areas where they can create value for the customer through this type of innovation. They can also build partnerships with other innovators and stakeholders by sharing knowledge and resources and experimenting with new digital technologies using the affordable loss principle [13].

2.2. Framework Development

AI adoption means that SME employees use AI technologies and free AI-based apps in daily business processes to improve business efficiency, productivity, and competitiveness cost-effectively to add customer value [11]. For example, a family-owned SME in the retail industry can use AI-powered tools to analyze customer data and provide personalized product recommendations. By adopting AI, SMEs can gain a competitive advantage, streamline operations, and enhance customer experience, ultimately leading to increased profitability and growth through AI-based innovation [7].

Adaptive resilience of family SMEs could refer to their ability to anticipate and respond effectively to changes in their business environment to deliver value to their customers [18]. Therefore, we can assume a positive relationship exists between AI adoption intentions and SMEs’ adaptive resilience [7,18]. Adopting AI is seen as a mechanism that can enhance the ability of SMEs to adapt to changing environments and remain resilient. DOI Web 2.0 posits that SMEs’ adoption of new technology follows a predictable pattern [5]. However, in the case of AI adoption, SMEs may be among early or late adopters, depending on their level of technological readiness and perceived benefits of AI adoption. Therefore, we can hypothesize that adopting AI may enhance SMEs’ adaptive resilience by providing them with tools to analyze and respond to changing market conditions [12]. The effectuation theory suggests that family SMEs could use cognitive and behavioral strategies to create new markets and opportunities rather than just reacting to existing conditions [3,13]. In AI adoption, the effectuation theory suggests that SMEs can use AI to create new products or services, enter new markets, or collaborate with other firms to create innovative solutions. In addition, by adopting AI, SMEs can leverage their existing resources and capabilities to create new value propositions and enhance their adaptive resilience [6,18,19]. Therefore, we have postulated the following hypothesis.

Hypothesis 1 (H1).

AI adoption positively influences adaptive resilience.

The Internet of Things (IoTs) has the potential to revolutionize industries and improve our daily lives with increased efficiency and convenience for all types of SMEs [20]. IoT refers to network devices that could be used and objects embedded with sensors, connectivity, and software that allow companies to exchange data and connect to the internet, which could be used to add customer value [8]. IoT edge devices can reduce latency and improve efficiency because data are processed at the edge rather than sent to a central server or the cloud, lowering the cost of real-time data analysis. With IoT-enabled devices, SMEs can automate processes, reduce costs, and improve productivity. For example, SMEs use sensors to monitor equipment performance and predict maintenance needs. However, while IoT edge can benefit SMEs, potential threats must be considered, e.g., the risks of cyberattacks [8]. IoT edge devices may have vulnerabilities that hackers can exploit, resulting in data breaches, the loss of sensitive information, and financial losses [21]. Regarding the threat of IoT edge for SMEs, DOI based on Web 2.0 suggests that SMEs can leverage existing networks and partnerships to accelerate the diffusion of IoT edge technologies [15,16,22]. However, it could result in customer data hacks. Effectuation theory also suggests that SMEs may be well-suited to adopting IoT edge [13], as they may be more agile and able to take risks than larger organizations but remain exposed to cyberattacks. So we can safely assume that the threat to IoT edge in the SME’s context could negatively impact adaptive resilience. Previously we have discussed how AI adoption could positively affect adaptive resilience. However, as SMEs using IoT edge are more vulnerable to cyberattacks [8,21], thus we can safely postulate the following hypothesis:

Hypothesis 2 (H2).

Threats to IoT edge negatively moderate the relationship between AI adoption and adaptive resilience.

Studies propose a positive relationship between the intention of SMEs to adopt AI and their level of digital innovation [4,19]. The DOI Web 2.0 theory [15,22] suggests that adopting innovations follows a pattern where innovators are the first to adopt, followed by early adopters, the early majority, the late majority, and finally, laggard SMEs. On the other hand, the effectuation theory proposes that SMEs use a unique approach [15,22]. For instance, rather than starting with a specific goal and identifying the means to achieve it, SMEs that use the effectuation theory say that any expert entrepreneur begins with the available resources and creates goals based on what they can achieve with those means for digital innovation [13]. The combination of these two theories suggests that the intention of SMEs to adopt AI can be positively influenced by early SME adopters who use an effectuation approach to decision making for digital innovation [6]. These early adopters can serve as role models for others in their social networks and demonstrate the benefits of AI adoption. Hence, we can hypothesize the following:

Hypothesis 3 (H3).

AI adoption positively affects digital innovation.

As stated earlier, IoT edge devices can reduce and improve business process efficiency because data are processed at the edge rather than being sent to a central server, lowering the cost for SMEs under DOI Web 2.0 [17,23] and effectuation theoretical assumptions. However, IoT edge devices are vulnerable to data breaches and the loss of sensitive information of customers and its firm [8,20,21]. Therefore, the threat to Edge IoT could discourage digital innovation practices. Thus, any unsafe IoT edge could weaken the relationship between AI adoption and digital innovation practices in the SME’s context. Thus, we can safely postulate the following hypothesis:

Hypothesis 4 (H4).

Threats to IoT edge negatively moderate the relationship between AI adoption and digital innovation.

Family-owned SMEs that can anticipate and respond effectively to changes and disruptions in their business environment can innovate digitally [3,18]. Therefore, this approach allows SMEs to position themselves for long-term success in an increasingly digital and competitive business environment [24]. Therefore, we have suggested that resilient SMEs are better equipped to respond to digital disruption and identify market opportunities [12]. For instance, an adaptive-resilient SME may be more likely to pivot its business model in response to changes in customer demand or adopt new digital technologies to improve its digital operations in the region.

Across the globe, IoT edge is equipped with connectivity and relates to interconnected network devices equipped with connectivity to gather, collect, share, and transfer customer data of business significance [20]. Thus, IoT edge development is essential for a family-owned SME. Many SMEs could encounter challenges, such as threats to IoT edge due to cyberattacks by competitors and larger firms [6,8]. Consequently, SMEs with more significant concerns regarding IoT edge threats are more likely to stick with the technology they are used to rather than switching to some new technology [21]. In the previous section, we also argued that IoT edge threats could weaken the positive relationship between AI adoption and the adaptive resilience of SMEs. Thus digital innovation practices could also suffer [4]. Nevertheless, once managers of family-owned SMEs feel that the IoT faces threats from competitors or stakeholders, the indirect positive relationship is weakened. Thus, we hypothesize the following:

Hypothesis 5 (H5).

Adaptive resilience positively affects digital innovation.

AI adoption among family-owned SMEs focuses on improving an SME’s adaptive resilience to explore market opportunities and cope with any emerging risk in online activity causing financial or reputation loss [12]. Lately, scholars integrating DOI with Web 2.0 [15,16] and effectuation theories have also acknowledged that AI adoption is a prerequisite for enhancing a firm’s resilience, which helps firms understand risks and gives SMEs strength for dealing with hardship and adversity [4,18]. Advancements in AI also demonstrate how the firm’s ability to anticipate and effectively respond to changes in its business environment could increase digital innovation practices in a given SME [9]. In previous hypotheses, we conceptualized how SMEs with higher adaptive resilience could interact with digital innovation practices. Thus, the latest cost-effective AI advancements could increase SMEs’ adaptive resilience, which helps them to focus on innovation processes and design new products and services that help them grow and obtain maximum profit by fostering digital innovation. Hence, we postulated the following (Figure 1):

Hypothesis 6 (H6).

Adaptive resilience mediates the relationship between AI adoption and digital innovation.

3. Methodology

We adopted the definition of family firms based on the theoretical perspective of socioemotional wealth and the FIBER Model (“Family Control, Identification, Binding Social Ties, Emotional Attachment, and Renewal of the Family Bond”) [2]. More specifically, we filtered those family-owned firms with a 50% share with family members and at least two employees employed with the family firm.

3.1. Sample and Data

Our empirical analysis was based on a dataset gathered by polling family-owned German businesses between March and May 2023. We randomly chose all the companies using the “Bureau van Dijk’s AIDA” database out of 72,031 German businesses. We determined that 23,000 family businesses fall within the definition of family firms. We emailed the technical personnel of these family SMEs. Most of them either did not respond or declined to participate in our survey. We obtained responses from 195 privately owned family businesses, but only 99 were acceptable since we rejected responses that left crucial issues unanswered. This equates to a response rate of 4.11%. In addition to our theoretical support for defining family firms, the SME definition of the “Institut für Mittelstandsforschung” Bonn, which requires the SME to employ fewer than 500 staff and have an annual turnover of less than EUR 50 million, led us to exclude those German firms that did not meet this criterion as German family-owned SMEs.

3.2. Measures

A five-point Likert scale was carried out for all the study measures (“where 1 = strongly disagree and 7 = strongly agree”) for each item: AI adoption, IoT edge, adaptive resilience, and digital innovation.

The scale of digital innovation consists of four items, adopted from a recent study [19]. A seven-item scale of AI adoption was also borrowed from a recent study [9], and a three-item scale of internal threats to IoT edge was adopted from the literature [20] and tested by recent explanatory research [8]. A five-item scale for adaptive resilience was adopted from a study [18]. Initially, the construct of resilience is discussed as organizational, psychological, and entrepreneurial resilience in the previous studies (see, e.g., [12]).

Controls: We applied the control variables in line with the previous study [3]. The first control was a business situation that asked respondents to reply with a good or bad business situation. Then we controlled for the four German regions in which family-owned SMEs were operating. We also controlled the industry in which the family-owned SMEs were operating.

4. Results

This research uses a professional version of SmartPLS 4.01 to validate the accuracy of the modeled data for family-owned SMEs. Using this approach, the constructs can be seen, and the research process may be streamlined, allowing for the association indicators and variables to be measured and tested.

4.1. Descriptive Statistics

Table 1. Descriptive statistics.

Variable	Sample (n = 99)	Percentage
Job Titles
Manager IT/MIS	20	20.20%
Data Analyst	25	25.25%
AI Programmers/Developers	30	30.30%
Data Engineer	13	13.13%
Other Tech Staff	11	11.11%
Size of Family SMEs
Small (11 to 50 employees)	47	47.47%
Medium (51 to 250 employees)	35	35.35%
Micro (up to 10 employees)	15	15.15%
Large (over 250 employees)	2	2.02%
Region
West Germany	45	45.45%
North Germany	30	30.30%
East Germany	14	14.14%
South Germany	10	10.10%
Sector
Production	40	40.40%
Services	40	40.40%
Retail	17	17.17%
Construction	2	2.02%
Business Situation
Good	61	61.62%
Bad	11	11.11%
Prefer not to say	27	27.27%

presents the descriptive statistics of respondents for individual- and organizational-level statistics. We collected data from five types of respondents related to AI or IT jobs. For instance, our respondents were IT/MIS managers (20.20%), data analysts, AI programmers/developers (25.25%), data engineers (13.13%), and other technical staff (11.11%). We collected data from small (47.47%), medium (35.35%), macro (15.15%), and large firms (2.02%). These family-owned SMEs were regionally dispersed across the west (45.45%), north (30.30%), south (10.10%), and east (14.14%). Most of the German family-owned SMEs were operating in production and services and reported that the business situation was good (61.62%).

4.2. Outer Model Assessment

Table 2. Standardized loadings, AVE, CR, and alpha.

Description	Items	Estimate	AVE	Alpha	CR
Digital Innovation (Dependent)	DI1	0.854	0.739	0.882	0.883
	DI2	0.844
	DI3	0.900
	DI4	0.838
AI Adoption (Independent)	AIA1	0.702	0.688	0.923	0.929
	AIA2	0.787
	AIA3	0.855
	AIA4	0.880
	AIA5	0.861
	AIA6	0.892
	AIA7	0.815
Threat to IoT Edge (Moderator)	EIoT1	0.870	0.750	0.835	0.855
	EIoT2	0.836
	EIoT3	0.891
Adaptive Resilience (Mediator)	DR1	0.841	0.774	0.926	0.929
	DR2	0.941
	DR3	0.903
	DR4	0.869
	DR5	0.840

describes the average variance extracted (AVE), Cronbach’s alpha, and composite reliability (CR) and the standardized factor loadings of items of four variables, including digitalization innovation (dependent), AI adoption intention (independent), IoT edge (moderator), and adaptive resilience (mediator). All the item loadings were above the threshold of 0.70, which makes our measurement tool fit for further testing of direct and indirect pathways [25]. All the AVE values for variables including digitalization innovation, AI adoption intention, IoT edge, and adaptive resilience were above the threshold level of 0.50, i.e., 0.739, 0.688, 0.750, and 0.774, respectively. Like the AVE values, all constructs’ CR values exceeded the threshold level of 0.70. Moreover, the values of Cronbach’s alpha exceeded the threshold level of 0.70, i.e., 0.882, 0.923, 0.835, and 0.926. Finally, the CR values for digitalization innovation, AI adoption intention, IoT edge, and adaptive resilience were 0.883, 0.929, 0.855, and 0.929, respectively. Please note that the survey items are provided in Appendix A of this study, and the German-translated version is available on demand.

Table 3. Discriminant validity.

	Variables	1	2	3	4	5	6	7
1	AI Adoption	1.000
2	Adaptive Resilience	0.714
3	Business Situation	0.131	0.192
4	Digital Innovation	0.617	0.674	0.198
5	IoT Edge	0.260	0.391	0.191	0.420
6	German Industry	0.069	0.064	0.134	0.170	0.069
7	German Region	0.062	0.061	0.057	0.172	0.205	0.09	1.000

shows the findings of an investigation into the discriminant validity. It is assessed by comparing the AVE’s square root with the constructs’ correlations and the correlation criterion’s heterotrait–Monotrait (HTMT) ratio. Cronbach’s alpha, composite reliability, and -values all fall within acceptable ranges, showing that all four constructions in our investigation (i.e., AI adopting, IoT edge, AR, and DI) have high internal consistency [26].

4.3. Direct Influences

In this study, we examined the three direct effects (

Table 4. Hypotheses testing.

Paths	$\mathsf{\beta }$	S.E	t-Value	p-Value	VIF	R2	Result
H1: AIA → AR	0.627	0.091	6.883	0.000	1.059	0.523	S
H1m: IoT Edge x AIA → AR	−0.217	0.113	1.929	0.027	1.002		S
H2: AIA → DI	0.295	0.131	2.254	0.012	1.887	0.483	S
H2m: IoT Edge x AIA → DI	−0.012	0.123	0.098	0.461	1.157		NS
H3: AR → DI	0.345	0.119	2.903	0.002	2.105		S
H4: AIA → AR → DI	0.216	0.073	2.960	0.002			S
Controls
BS → DI	0.083	0.288	0.288	0.387	1.081
GI → DI	−0.16	0.067	2.397	0.008	1.061
GR → DI	0.129	0.074	1.732	0.042	1.077
Moderated Mediation:
IoT Edge x AIA → AR → DI	−0.075	0.047	1.608	0.049			S

Note: Artificial intelligence adoption (AIA), Internet of Things (IoT) edge, adaptive resilience (AR), digital innovation (DI), business situation (BS), German industry (GI), German region (GR), supported (S), not supported (NS).

) of AI adoption among German family-owned SMEs on adaptive resilience (H1: AIA → AR), AI adoption on digital innovation (H2: AIA → DI), and adaptive resilience on digital innovation (H3: AR → DI). The analysis shows that statistical evidence exists to support a statistically significant and positive influence of AI adoption on AR (H1: AIA → AR = 0.627, t = 6.883, p < 0.001), influence of GKA on OR (H2: AIA → DI = 0.398, t = 2.254, p < 0.05), and influence on OR (H3: AR → DI = 0.345, t = 2.903, p < 0.001).

4.4. Moderating and Mediating Effects

This study also examined the two moderating effects of IoT edge and the mediating role of adaptive resilience (Table 4). We could find statistically significant evidence for the first moderation (H2m: IoT Edge x AIA → AR = −0.217, t = 1.929, p < 0.05). However, our second moderation was not substantiated (H3m: IoT Edge x AIA → DI). Therefore, we can say that the IT team of family SMEs feels that there is a substantial threat of using IoT edge technology, and larger firms can time attach the small firm to take sensitive customer and financial data.

Finally, our mediating role of adaptive resilience was substantiated for family-owned SMEs (H4: AIA → AR → DI = 0.216, t = 2.960, p < 0.05). This confirms that AI adoption indirectly causes organizational resilience through adaptive resilience among family-owned SMEs to support the mediating role of AR (Table 4).

5. Discussion and Conclusions

We conclude that family-owned SMEs intend to adopt AI to improve digital innovation and enhance organizational resilience. Still, family-owned SMEs face challenges using IoT edge and perceive IoT edge as a threat. The results are partially consistent with previous studies (e.g., [3,4,11]).

5.1. Contributions

Our study has contributed in multiple ways. First, we have contributed by integrating theories of two disciplines. For instance, we used the effectuation theory (ET), a powerful tool for family SMEs seeking to adopt AI technologies, drive digital innovation, and build adaptive resilience [13]. Theoretically, we have added to the body of knowledge by showing how the effectuation theory can be used in family-owned SMEs to connect AI usage and the possible risks of using Internet of Things (IoT) edge devices. Concerning AI adoption, we tested and explained that the effectuation theory’s focus on making the most of existing resources and networks could help family-owned SMEs integrate AI technologies successfully. Additionally, by recognizing and using their unique strengths and relationships, family-owned SMEs can find AI apps that fit with their skills and goals, giving them a competitive edge and making it easier for them to use digital innovation to gain market share. However, some risks come with the use of IoT edge. Family SMEs increasingly connect their businesses to IoT edge devices, so they must deal with possible hacking risks. If these IoT gadgets are not properly locked down, they can be used for unauthorized access, data breaches, and other nasty things. Therefore, as family SMEs grow, they should put in place robust security measures and find the money to invest in the latest technology, such as encryption protocols and frequent vulnerability assessments, to protect their operations and sensitive information from these threats. Therefore, the effectuation theory can help family SMs adopt AI technologies by using their existing resources and pointing out how important it is to address security concerns when adding IoT edge devices into their operations. Second, we used the diffusion of innovations Web 2.0 theory, which can help family-owned SMEs figure out how to use AI and deal with the possible risks of using IoT edge devices. First, DOI Web 2.0 emphasizes how vital communication and social networks are for getting people to use new ideas [4,15,22]. We looked at and described how family-owned SMEs could use DOI to encourage the use of AI by making it easier for their stakeholders, including employees and partners, to share knowledge and create a creative environment for digital innovation. Second, we showed how vital trialability and observability are by using DOI. Family-owned SMEs can reduce the danger of IoT edge devices by testing them in controlled environments, keeping an eye on how well they work through tech partners at a reduced cost, and asking customers for feedback on potential security flaws. This will make it easier for the devices to integrate safely into their operations. Third, our contribution is contextual [1,14,27,28]. We brought unique evidence from family-owned SMEs operating in four key regions. Scholars studying how SMEs with family ownership in Germany use AI for digital innovation can learn a lot from the situation in Germany [3,7,29]. In Germany, family-owned small and medium-sized enterprises (SMEs) are known for their focus on the long term, loyalty to quality, and close ties to their local communities. Therefore, we added to this distinctive context by studying how these family-owned SMEs use AI for digital growth.

5.2. Limitations and Future Research

Similar to previous studies, this one contains limitations that require additional investigation. The study’s sampling of family-owned businesses in Germany limited generalization only to small family firms. Moreover, no comparison is produced. Therefore, in the future, family business scholars can replicate this framework in other European nations to generalize this framework for family firms. Refer to Table 4. Future research is possible to develop a conceptual framework and test moderated mediation. Additionally, the moderating roles of the German industry (GI) and German region (GR) can be tested. Moreover, the research can collect data from larger family firms and see if these larger family firms have more planned resilience compared with smaller family firms [18,30,31].

5.3. Practical Implications

For the IT team of family-owned SMEs, adopting AI and avoiding threats from IoT edge devices have strengths and weaknesses. First, HR should help the IT team learn new skills to use AI. For example, the IT team needs to improve AI technologies such as machine learning, natural language processing, and data analytics to integrate AI successfully into the organization’s processes and systems for digital innovation and make the organization more resilient in times of crisis. Second, AI needs a lot of different kinds of high-quality data. To support AI algorithms and models, the IT team should ensure that data governance policies, data integrity, and data security are in place. Third, when it comes to threats to IoT edge, the IT team should put in place strong security measures for IoT edge devices, such as safe authentication, data encryption, and regular firmware changes. By dividing the network into sections that keep IoT edge devices away from vital systems and sensitive data, security breaches can have less effect. Fourth, IoT edge devices should be constantly monitored for weaknesses or strange behavior. The IT team should quickly install security fixes and firmware updates to reduce risks. Lastly, the IT team should offer programs and training to teach workers about AI technologies, IoT edge devices, and the risks that come with them. This helps family-owned SMEs build a security mindset and responsibly use AI. Overall, the IT team is one of the most important parts of adopting AI and dealing with the risks that come with IoT edge devices. Their knowledge of AI technologies, data management, security protocols, and compliance is essential for building a strong and safe environment for family-owned SMEs to use AI’s benefits while doing digital innovation and creating adaptive resilience.