2 min

Tags in this article

, , ,

Most global organizations do not have the necessary resources to accommodate artificial intelligence (AI) and machine learning (ML) initiatives. This information comes from a new survey conducted by Rackspace Technology.

After polling 1,870 IT pros, the report shows that 82% of organizations worldwide are yet to introduce infrastructure that makes it easy to implement AI and ML.

Rackspace says that the failure to implement can be accounted for by various factors that include:

  • The lack of quality data (34%)
  • Skills gap (34%)
  • Poor strategy conception (31%)

With these obstacles in place, getting any organization started on AI and ML becomes quite difficult.

The inevitable future

There is another problem these organizations face. Many of them are unsure whether they would like to outsource their AI and ML programs or build everything in-house from scratch.

According to Tolga Tarhan, the Chief Technology Officer at Rackspace Technology, IT decision-makers turn to AI and ML to improve customer satisfaction rates and efficiency in business processes in almost all industries.

It is becoming clear that with the emergence of modern business intelligence, data analysis is best left to AI and ML models to find out the best decisions to make.

If you are going to AI/ML, do it properly

Tarhan added that before organizations take on any machine learning or artificial intelligence initiative, they must undergo a thorough cleaning of data and data processes. The message here is to ensure that the data fed into AI or ML models is the right data.  

The data cleaning and processing has to be done reliably and cost-effectively.

Rackspace says that the establishment of clear KPIs is important in measuring the success of the initiatives. Commonly used KPIs include profit margin, revenue growth, and customer satisfaction/net promote scores.