The advent of generative AI gives organizations more levers to drive growth for their customers, writes Greg Taylor, vice president at Databricks.
Driven by the need for efficiency, productivity and innovation, the adoption of artificial intelligence (AI) in business has accelerated in recent years. Recent developments in generative AI are further accelerating the adoption of the technology, with analyst firm IDC predicting that spending on AI systems in Australia will reach $3.6 billion by 2025.
That makes sense: the local business community is clearly aware of the potential efficiency gains, cost savings and long-term promise that AI technology can deliver.
For example, by delegating time-consuming, repetitive tasks to efficient AI systems, companies can free up time for people to focus on higher-value tasks like creative strategic thinking and innovation.
So it may come as a surprise that only 30% of respondents to an MIT Technology Review Insights survey commissioned by Databricks said their organizations are rapidly adopting AI tools. Much more (42%) said their organizations are moving at a moderate to very slow pace when it comes to adopting AI tools.
While 98% of Australian businesses are using some form of generative AI, only 30% have actually adopted it, meaning the business community still has a long way to go to make the most of such technology. This AI gap creates new opportunities for corporate partners, such as consulting firms and systems integrators, to leverage AI to drive greater efficiency and effectiveness for their customers.
References: V2 Digital's State of AI in Australia: Embrace AI or get left behind.
Efficiency and cost savings
The primary use cases for AI among enterprise customers are varied but include augmented customer service agents, digital assistants, smart innovation, automation, etc. In the current economic climate, one of the major focuses is cost control through increased efficiency, and AI has the potential to significantly improve efficiency.
For example, AI-powered automation (in this case, so-called “boring AI”) can help streamline monotonous and repetitive tasks like data entry, invoice processing, reconciliation, etc. Delegating these time-consuming and often manual tasks to efficient AI systems frees up employees' time to focus on higher-value tasks like problem-solving, strategy, and innovation.
It's no surprise, then, that the majority (70%) of those surveyed by MIT Technology Review Insights see cost reduction as a top priority for AI systems. Against this backdrop, four in five Australian businesses (80%) expect to become at least 25% more efficient over the next two years thanks to AI technology.
Across Australia and overseas, there is a shortage of the skills needed to implement advanced AI technologies in-house, so businesses are increasingly turning to external consulting and integration partners to provide a viable AI strategy and the know-how to implement it effectively. This is where consulting firms have an opportunity to step in and help.
Leveraging Enterprise Data for AI Edge
Implementing an in-house AI system for a specific business outcome within your enterprise is quite different from leveraging an external internet-based platform like ChatGPT to quickly and easily create marketing content. Because data underpins the effectiveness of AI-based solutions, it is important that applications that rely on enterprise data have easy access to it.
However, incorporating business data, which may contain customer personal or commercially sensitive information, into a publicly available AI platform can pose potential risks. AI applications that leverage enterprise data are best deployed within an enterprise's own IT estate, but for this to work, enterprise data must be properly managed.
Until now, it has been difficult to properly manage data for AI applications that require large datasets to work properly, which means a large part of the role of external partners helping companies implement transformative AI in-house is not only the design and implementation of the AI models themselves, but also the data management platform that underpins them.
Fortunately, a new generation of data intelligence platforms is emerging that are making complex in-house AI applications accessible to many companies. Based on the data lakehouse architecture pioneered by Databricks, these data intelligence platforms use AI models to deeply understand the semantics of enterprise data, making it easier to understand data across the enterprise.
Using AI to power AI
Adopting an AI-powered data intelligence platform can give companies the level of insight and visibility they need to properly train and operate their AI models, especially models like ChatGPT that require very large data sets to work as expected. For external partners helping companies implement AI solutions, this means achieving this through a two-fold approach.
For any AI initiative to be successful, you must first build a data intelligence platform and ensure that your client's data is available. Only then can the second part happen: training the AI model. This point is important: the quality of the data you use to train your AI model directly impacts the effectiveness of its application and the business outcomes it delivers.
AI-powered data intelligence platforms enable essential capabilities for AI solutions, such as semantic cataloging and discovery, natural language access, automated management and optimization, and enhanced governance and privacy, which is one of the reasons why such data platforms open the door for external enterprise partners to delve deep into new AI use cases and bring new levels of efficiency and innovation to client organizations.
Leveraging a data intelligence platform to drive AI not only improves our enterprise clients’ AI capabilities, but also potentially improves overall data utilization by making organizational data more usable for other functions such as general insights, analytics, predictions, and enhanced decision-making, strengthening our clients’ businesses.
