Riverbed’s Fernando Castanheira advises companies not to run before they can walk when it comes to AI implementation.
With artificial intelligence (AI) technologies ramping up all the time, implementing AI is now viewed as essential for staying competitive, boosting productivity and driving innovation.
Despite the clear promise, recent findings from Riverbed’s Global AI & Digital Experience Survey reveal that many business leaders are still making fundamental errors in their approach to AI integration. These mistakes must be avoided to ensure that AI strategies can evolve from a solid and healthy foundation. After all, you can’t run before you can walk.
Overconfidence in AI capabilities
One of the most prevalent issues is a widespread overestimation of AI’s readiness to deliver results, which mainly stems from a combined lack of readiness, skills and dedicated resources; an absence of industry-comparable use cases; insufficient internal data quality; and a broader misunderstanding of AI’s capabilities.
In fact, a staggering 82pc of business leaders across all industries believe they are ahead of their competitors when it comes to AI, but only 37pc say they are truly prepared to implement projects right now. Unbridged, this gap between perception and reality could cause project setbacks and misallocated resources, which would result in the breakdown of stakeholder trust and, ultimately, major challenges in AI deployment.
To be more realistic with their expectations, organisations should conduct a thorough AI readiness assessment – ensuring that they are prepared with accommodating internal infrastructure, accurate data quality and a culture of continuous learning before pushing forward with large-scale AI initiatives.
Underutilisation of internal data
Forbes has described data as the “lifeblood” of AI, because the “quality and quantity of the data it ingests are paramount to its effectiveness”. Yet, as it stands, many organisations are failing to harness their own internal data. While 85pc of leaders acknowledge that great data is critical for AI success, only 34pc rate their data as excellent for consistency and standardisation.
This discrepancy points to a missed opportunity: enterprises are sitting on an abundance of data that could be used to power their AI, but they aren’t leveraging it properly across all relevant areas of the business. Or their data fundamentally lacks the accuracy and reliability to consistently support AI in the first place.
To standardise data, organisations should invest in enhanced data management tools and develop a comprehensive strategy for data collection, storage and standardisation across every department that needs it.
Cross-departmental data-sharing, collaboration and consistency across datasets for uniform analysis will ensure reliable and repeatable results from new AI models, thereby building trust in the data-driven decisions they produce and delivering better business outcomes in the process.
Cybersecurity and compliance roadblocks
AI’s rapid integration into the world of work promises many benefits, but it does unfortunately introduce a whole new arena of risk management. Namely, malicious digital threats such as data breaches and AI-targeted ransomware, as well as formal pressure from compliance regulations such as GDPR, ISO and NIS2.
Many business leaders are cautious about AI adoption for exactly these reasons, with 43pc hesitant to invest more due to cybersecurity concerns, and 36pc due to regulatory challenges. Clearly, the emergence of AI could potentially expose sensitive information if not adequately protected – so much so, that McKinsey also highlighted that 53pc of organisations acknowledge cybersecurity as a generative AI-related risk.
Rather than letting these apprehensions stall progress, businesses should choose to view them as opportunities to broaden their AI strategies. Investing in secure AI architecture and factoring in compliance from the ground up can prevent similar roadblocks in the future. Addressing these challenges early means AI projects become not only innovative, but also resilient to both industry regulations and internal security needs.
Underestimating AI’s role in enhancing experience
A substantial 94pc of leaders agree that AI can help them to deliver better digital experiences for their end users. This can entail seamlessly personalised interactions, faster response times and the fostering of an engaging, highly accessible environment across all digital platforms.
But, in reality, this transformative potential is still often overlooked.
AI can make these aspirations achievable in practical ways. For instance, intelligent service desk solutions enhance customer support by using built-in AI functionality to automate inquiries and remediate issues – enabling continuous and accurate support with minimal human intervention.
By leveraging data that’s dynamically collected from user experiences, AI is also able to tailor user-specific recommendations, making each interaction feel frictionless. Doing so will not only enhance AI operations, but also help organisations stand out as being technologically user-friendly in a highly competitive market.
Lack of oversight in IT environments
The seamless integration of AI depends on a robust and regulated IT infrastructure. However, many organisations are inadvertently guilty of lack of oversight in this area, especially in complex IT environments that span cloud, hybrid and multidevice systems. This obscurity of vision can cause difficulty in monitoring performance, challenge data security and consistency, as well as increasing the risk of operational downtime.
Decision-makers should therefore regularly audit their infrastructure for weaknesses, while still looking to implement centralised management solutions which provide real-time monitoring across the end user, network, data and applications. These tools offer real-time insights and automated responses to the health and performance of a complex IT environment and support teams with their AI initiatives.
Also, the power to transform oversight into foresight allows for the quicker identification and resolution of issues, eliminating the risk of AI project challenges due to system vulnerabilities.
One step at a time
AI holds immense potential to transform the future of business operations, but only if organisations make strides towards avoiding these common mistakes. To build a healthy framework for AI, businesses must therefore be realistic about their readiness; improve their data quality; familiarise themselves with compliance standards; adapt to new user expectations; and clarify insights into their own IT environments.
By taking the right approach and implementing practical AI that works and scales, organisations can reach new commercial and operational heights, delivering a significant competitive advantage and improved business outcomes.
By Fernando Castanheira
Fernando Castanheira is Riverbed’s global CIO responsible for all aspects of IT including digital transformation, cyber/IT security, infrastructure, operations and business platforms. Prior to joining Riverbed, he worked at JPMorgan Chase in various senior IT roles, most recently as a managing director and CTO.
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.
