Data collection errors, inconsistent data formatting issues across vendors, data storage issues, and network monitoring blind spots were the top issues that are impacting this data quality.
Bad data leads to bad AI insights. Network teams will need to assess their data before they invest time and money in AI tools. If you’ve identified data quality issues in your organization (and most of you probably have), raise them with the vendors that are trying to sell you AI solutions. Discuss how those data problems might impact the value their AI tools and what steps you can take to remediate those issues.
Teach yourself to evaluate AI
Only 39% of the IT professionals EMA surveyed were completely confident in their organization’s ability to evaluate the AI-driven network management technology. Success with AI correlated very strongly with confidence.
EMA isn’t advising clients to become AI experts, capable of developing and debugging algorithms. However, IT organizations should take vendors at their word that their AI solutions will deliver value.
EMA research found that successful organization are leveraging continuous AI accuracy monitoring and feedback loops as they start a proof of concept and proceed to production. They also have in-depth conversations with their vendors about how those vendors train their AI models. This latter approach can help a you determine whether that training might fit with the type of network you manage. Finally, look for vendors that leverage explainable AI tools that make it clear how the technology derives insights from data.
Foster trust among NetOps personnel
AI can deliver value only if the networking team is using it. Unfortunately, only 31% of IT professionals told EMA that they completely trust their AI tools. This trust correlates very strongly with how often network teams actually use AI tools.
