Recently, Mark Colaluca, Senior Vice President at HCLTech, sat down with Gary Zhao, Founder of Digital Innovation Valley and an HPE strategic advisor, to explore how Digital Innovation Valley has navigated its AI-driven transformation. Their conversation highlights critical obstacles, the crucial role of strategic partnerships and the importance of increased collaboration between government and private sector, offering a blueprint for organizations aiming to thrive in the AI age.
Navigating unexpected obstacles in AI adoption
Despite extensive preparations prior to the impact of modern AI, Digital Innovation Valley encountered several hurdles once foundational models and advanced analytics became widely accessible.
1. Talent shortage and development gaps
“We made extensive preparations,” explains Zhao, “but after AI emerged, our talent cultivation could not meet market expectations.” Although Digital Innovation Valley had invested in upskilling programs, the demand for AI experts, particularly data scientists and analytics engineers, outpaced the pace of traditional training pipelines. Cultivating professionals with diverse competencies across machine learning, deep learning and data governance proved to be a lengthy process.
Even dedicated partnerships with universities and other vocational institutes have yet to fully close this gap, underscoring the urgency for accelerated talent development initiatives and industry-academia collaborations.
2. Reinventing work patterns and organizational structures
Drawing a parallel with the dawn of electrification, Zhao notes that “when electricity was born, it took nearly a century for widespread practical application because entire work patterns had to be fundamentally transformed.” In today’s AI era, companies must rethink organizational hierarchies, decision-making processes and cross-functional workflows. Legacy teams structured around siloed departments need to evolve into agile, multidisciplinary squads that can iterate quickly on AI pilots. This cultural and structural shift requires steady, continuous investment in change management, as well as a willingness to tolerate short-term disruption in service of long-term innovation.
3. Delivering rapid, low-cost value
Even when AI proofs of concept succeed internally, ensuring that clients realize tangible and cost-effective benefits remains a major challenge. “How do we help our clients obtain tangible, rapid, low-cost benefits through AI?” asks Zhao. Without clear metrics, well-defined key performance indicators (KPIs) and a focus on minimizing deployment complexity, AI projects often stall or produce marginal returns. To mitigate this risk, Digital Innovation Valley has begun adopting modular AI solutions, sequenced so that each incremental step yields measurable ROI, to foster client confidence and reduce the time to value.
The power of strategic partnerships
From high-level strategy to specialized technological integration, external partners have been instrumental in accelerating Digital Innovation Valley’s AI roadmap. Zhao categorizes these alliances into two main types:
Consulting-led alliances
HCLTech provides holistic guidance within a global ecosystem, helping to address key business challenges and shape foundational products. For instance:
- Low-carbon power-utility solutions: Digital Innovation Valley sought to develop China’s first power-bidding system with a focus on carbon-reduction goals. Drawing on HCLTech’s global experience in low-carbon IT implementations, the partners co-developed a foundational platform that integrates advanced analytics with real-time grid data
- Global best practices integration: Through consulting workshops, HCLTech experts helped Digital Innovation Valley benchmark against international standards for utility bidding, ensuring that the solution would be both scalable and competitive on a worldwide stage
This consulting-led approach enabled Digital Innovation Valley to navigate regulatory complexity, validate critical assumptions and secure a “big win” in its very first power-bidding pilot — positioning the company as an industry frontrunner.
Technology-specialist collaborations
Beyond strategy, Digital Innovation Valley relies on a network of specialized technology providers to supply niche capabilities in domains such as text analytics, audio processing and large language modeling.
“Throughout this era, the division of labor has become increasingly subdivided,” observes Zhao. By tapping into each partner’s best-in-class technology, Digital Innovation Valley can assemble modular AI stacks, mixing and matching capabilities to solve concrete business problems without overextending internal resources.
Embracing the AI era and smart manufacturing
According to Zhao, “We’ve reached an age of AI where it’s integrated into our daily life and work. If you cannot embrace AI, you’re not aligned with the development trend.”
He traces AI’s rapid evolution:
- Data governance: Initial efforts focused on cleaning, consolidating and securing enterprise data. Many organizations spent years establishing data lakes, governance frameworks and basic analytics pipelines.
- Big and foundation models: With the arrival of deep-learning-based models (such as large language models like the GPT-series and emerging proprietary systems such as DeepSeek), AI capabilities have leaped forward, enabling more sophisticated natural language understanding, computer vision and predictive analytics.
- Future intelligent agents: Zhao predicts “millions of intelligences out there,” suggesting an eventual landscape where specialized AI agents handle end-user queries, automate cross-enterprise workflows and dynamically recombine services on demand.
For manufacturing and specifically smart manufacturing, this evolution could translate to:
- Predictive maintenance: AI models analyze sensor data to predict equipment failures, schedule repairs proactively and minimize unplanned downtime
- Quality control: Computer vision systems automatically inspect components for defects, reducing reliance on manual inspection
- Supply chain optimization: Advanced analytics forecast demand variations, optimize inventory levels and dynamically reroute logistics based on real-time data
By collaborating with HCLTech, Digital Innovation Valley has already identified practical entry points, such as integrating AI-driven image recognition into production lines and is actively demonstrating proof of concept in smart factory environments. As Zhao notes, “The ‘last mile’ of AI integration has been partly solved by players like Amazon, Facebook and Google. Now, the question is how to redefine internal work patterns to leverage these breakthroughs effectively.”
Government–private collaboration: Building an innovation ecosystem
Digital Innovation Valley offers a compelling example of how government and private enterprises can co-create an innovation ecosystem that accelerates technology adoption, supports startups and advances sustainability goals.
1. Industrial platform and resource integration
The local government builds an industrial ecological platform that aggregates resources, including land, funding and R&D facilities and opens demonstration zones. By “setting the scene,” it enables value-added enterprises to pilot AI and low-carbon solutions in a controlled, high-visibility environment.
2. Clear role assignment
Government entities identify and attract high-potential companies, such as HCLTech, with its 200,000+ professionals and global client base and invite them to anchor the ecosystem. At the same time, they attract specialized partners like Quantum Matrix (text analytics), top university research institutes and state-owned investment platforms, each with defined responsibilities.
What is the government's role in this model? “First, it ensures the introduction of industries. Second, it enables setting the scene. Third, it provides more investment platforms and investment opportunities,” says Zhao.
On the enterprise side, Digital Innovation Valley and partners like HCLTech bring global use cases, mature R&D methodologies and supply-chain experience. “HCLTech, for example, has so many global use cases and solutions, and we can apply these use cases to [create a cycle of knowledge transfer and practical demonstration,” continues Zhao.
Ultimately, this collaborative model accelerates the commercialization of emerging technologies, especially in AI, smart manufacturing and low-carbon initiatives, while ensuring alignment with regional sustainability objectives.
