For some time, artificial intelligence (AI) has been a disruptive force not only in the technology sector, but across numerous industries. And it has only grown more so after the mainstreaming of generative AI with the release of ChatGPT by OpenAI. Most technology companies are already creating their own generative AI services, and the great majority of organizations are investigating the opportunities and threats it presents. Traditional data analytics suppliers have been less loud, although disruption has been occurring for years before to the development of ChatGPT.
Let’s take a look at how artificial intelligence has been driving innovation and quietly changing the space.
The fundamentals of data analytics
Humans are producing an increasing amount of data, owing partly to the digitization of our society, ranging from databases storing standardized information about residents to user-created material on social media platforms and sensor data generated by smartphones and industrial gear. Many industry projections predict that over 175 zettabytes of data will be generated by 2025. As a result, we are drowning in data, and making sense of it all is increasingly harder.
For decades, the term ‘big data’ has been used to describe enormously vast, diverse data sets that, when aggregated, can show patterns, trends, and relationships, particularly those relating to human conduct and interactions. The five V’s of big data are high velocity, volume, value, variety, and veracity.
Data analytics enables us to transform raw data into usable insights and actionable knowledge. There are four main forms of data analytics:
- Utilizing historical data, descriptive analytics tries to explain what happened. Affinity grouping, association rule learning, principal component analysis, anomaly detection, and clustering are some of the methods that may be applied.
- For the purpose of determining why things occurred, diagnostic analytics builds on descriptive analytics. An additional investigation is conducted into anomalies and areas of interest. Statistical methods can be used to explain and gather more relevant data.
- Using statistical methods and machine learning algorithms, predictive analytics categorises upcoming events or predicts unknowable results. The likelihood of events, especially those not on an organization’s radar, is evaluated by analyzing historical and current data.
- Following predictive analytics, prescriptive analytics advises enterprises on their next course of action. Prescriptive analytics suggests ways to enhance results by spotting outliers and possible winners and losers. This stage of business analytics is frequently seen as being the last.
Not every data analytics is created equal
There is a possibility that not all data analyses are created equal according to their history and application of AI.
Machine learning (ML) techniques were employed in the initial wave of AI disruption to deliver data-driven suggestions by analyzing massive volumes of data and evaluating “what if” scenarios. The field of prescriptive analytics was improved and disrupted by this breakthrough.
Two distinct categories of providers of prescriptive analytics can be distinguished here. Traditional data analytics companies with their roots in descriptive analytics are included in the first group. These companies include SAS, IBM, Microstrategy, Oracle, and SAP. Many were established during the turn of the century, and others even earlier, in the 1970s, like SAS and Cognos (now a part of IBM).
The second group consists of AI-native suppliers like C3.ai, CognitiveScale, and H2O.ai that enable businesses use ML to automate operational decision-making. However, the first decade of the 2000s saw the emergence of this initial wave of AI disruption. For instance, C3.ai, H2O.ai, and CognitiveScale were all formed in 2009, 2011, and 2013, respectively. At the time, there were no indications of generative AI, and Alphabet would not create the Transformer model (the “T” in GPT-3 and ChatGPT) until 2017.
Though it’s possible that a third group has also entered the prescriptive analytics field with the development of generative AI. After consuming corporate data as a prompt, large language models (LLMs) from companies like OpenAI, Alphabet, Meta, Cohere AI, or Anthropic can efficiently offer strategic recommendations. The data analytics sector has a great chance to be disrupted, even though it is still in its early stages.
Prescriptive analytics, or generative AI, can be used to either improve visualisations or offer recommendations. Analytics companies are integrating generative AI functionality to automatically produce initial analysis and visualisation from simply a textual query on a dataset. For instance, ChatGPT and Microsoft PowerBI have been connected to examine data sets and automatically produce a summary analysis. With the release of Microsoft 365 Copilot, ChatGPT is now integrated with analytics software like Excel and PowerBI. It could be argued that AI has revived a data analytics market that was long on its way to maturity. In fact, according to GlobalData, the global market for data analytics will be valued $188.8 billion in 2027, representing a 13% compound annual growth rate from 2022 to 2027.
