Image generated by Deeptech Times using Google Gemini
In the early days of enterprise software, competitive advantage was often defined by the strength of the product itself. But as AI enters its next phase of enterprise adoption, that paradigm is rapidly shifting.
According to insights shared by Databricks’ Joseph Bosco, partnership manager for APJ, the future of AI platforms will not be determined by products alone, but by the ecosystems that surround them.
This evolution reflects a deeper truth about AI today: we are past the stage where a single clever model is enough. The real winners will be platforms that can pull together data, models and partners into a single, reliable and well-governed backbone for enterprise transformation.
From standalone AI to ecosystem-led platforms
The rise of AI platforms over the past decade has already simplified how organisations build and deploy machine learning models. Databricks, for instance, has helped redefine enterprise data architecture through its lakehouse approach.
Yet, as Bosco points out, the next phase of competition is no longer about who has the best model or the most features.
“Features can be replicated. Algorithms are not proprietary in the long term,” he notes. “What becomes defensible is the ecosystem.”
This marks a shift from product-centric thinking to ecosystem orchestration.
AI is no longer a standalone capability: it is a system of interconnected layers involving infrastructure, data, models, applications and domain expertise. No single vendor can own all of these effectively. Instead, success lies in how well these components are integrated and governed as a cohesive whole.
The strategic role of partners
Nowhere is this ecosystem-driven approach more evident than in APAC, where diversity across markets adds layers of complexity to AI deployment.
Bosco explains that while Databricks provides the underlying platform, it is the partner ecosystem that ensures AI solutions are successfully implemented on the ground. Partners help navigate local regulatory environments, modernise legacy data estates and build industry-specific use cases that are not only technically sound, but actually adopted by the business.
This is particularly critical in Southeast Asia, where each country operates under vastly different governance frameworks, data regulations and levels of digital maturity.
In this context, partners act as the bridge between global platforms and local realities. They bring contextual understanding, execution capability and domain expertise to turn abstract AI potential into tangible business value.
Developers as the engine of innovation
Beyond enterprise partners, developer communities form another critical pillar of the ecosystem. Databricks’ roots in open-source technologies such as Apache Spark, Delta Lake and MLflow reflect a long-standing commitment to building with and for engineers.
This developer-first approach is not just philosophical but also strategic. By investing in open technologies, practical training programmes and reusable templates, Databricks enables developers and partners to build and deploy solutions quickly without starting from scratch.
The result is a compounding effect. As more developers contribute to and build on the platform, the ecosystem becomes richer, more innovative and more resilient. This continuous cycle of contribution and adoption is what ultimately drives long-term platform success.
Open, extensible and secure by design
A key requirement for any ecosystem-driven platform is openness. However, openness alone is not enough. Enterprises also demand control, governance and security.
Databricks addresses this balance through a layered approach. At the data layer, technologies like Delta Lake ensure openness and interoperability, allowing organisations to store and access data in standardised formats. On top of this, governance frameworks such as Unity Catalog provide fine-grained access control, ensuring that data is used responsibly and securely across the organisation.
At the same time, capabilities like Mosaic AI enable enterprises to operationalise GenAI in a way that is aligned with security and risk requirements. This is particularly important as organisations grapple with the challenges of deploying LLMs and other advanced AI systems in regulated environments.
The combination of openness, extensibility and governance allows enterprises to innovate rapidly without compromising control.
Vertical AI: Where partners create differentiation
While Databricks positions itself as a horizontal platform, the real differentiation often emerges at the industry level and this is where partners play a pivotal role.
Bosco emphasises that sector-specific expertise cannot be easily standardised. The nuances of banking, telecommunications or retail require deep domain knowledge, from understanding regulatory requirements to modelling industry-specific risks and opportunities.
Rather than attempting to build these capabilities internally, Databricks relies on its ecosystem to deliver verticalised solutions. Partners take the core platform and layer on their industry expertise, creating repeatable accelerators that can be deployed across similar organisations.
This approach not only accelerates time-to-value, but also ensures that AI solutions are grounded in real-world business contexts.
Why ecosystems matter for Southeast Asia’s AI journey
For organisations in Southeast Asia that are just beginning their AI journey, the case for an ecosystem-driven approach is particularly strong.
Building an AI stack entirely in-house may seem appealing from a control perspective, but in practice, it is often slow, costly and difficult to scale. Bosco likens it to building a car from scratch: technically feasible, but rarely practical.
Instead, organisations can achieve faster and more reliable outcomes by leveraging established platforms like Databricks, combined with experienced partners. This allows them to stand up initial use cases quickly, demonstrate value and gradually build internal capabilities over time.
As the AI landscape evolves, so too does the composition of the ecosystem. In addition to traditional partners such as cloud providers and system integrators, new categories of players are emerging.
Model providers, data providers and specialised AI tooling companies are becoming increasingly important, contributing new capabilities and innovations to the ecosystem.
Databricks is positioning itself as a neutral platform where all of these players can integrate and collaborate.
This neutrality is key. By avoiding proprietary lock-in and enabling interoperability, the platform allows enterprises to choose the best tools and partners for their needs, rather than being constrained by a single vendor’s ecosystem.
From platforms to economies
Looking ahead, Bosco is clear that the next three to five years will see ecosystems become the defining factor in AI success.
The competitive edge will not belong solely to platforms with strong technical cores, nor to those with large partner networks in isolation. Instead, it will belong to those that successfully combine both: a robust, scalable platform at the centre, surrounded by a vibrant and engaged ecosystem.
This requires deliberate investment on multiple fronts: strengthening the core platform, nurturing developer communities, expanding partner networks and enabling new types of ecosystem participants.
In this sense, AI platforms are evolving into something far more complex than traditional software products. They are becoming ecosystems and, increasingly, entire economies of innovation.
