Future Predictions of AI Technologies
February 20, 2020

For many organisations AI is still a long way off in the distance, a technological fantasy of futuristic proportions. PWC’s 2020 AI Predictions report says that those businesses need to ‘reality check’ now and start laying the groundwork for an ‘AI-powered future’ or risk being left behind. In fact, in their third annual report PWC findings show that only 4% of executives surveyed plan to deploy AI enterprise-wide in 2020, when a year ago approximately 20% were planning to.

So what happened? Why the significant drop in confidence? PWC suggest that the need to ‘focus on the fundamentals before enlarging AI projects’ is now the focus.

According to the Gartner 2020 CIO Agenda Survey, leading organisations are preparing to to double their number of artificial intelligence (AI) projects over the next 12 months, with over 40% planning to actually deploy AI solutions by the end of 2020. Like PWC, Gartner reports that the reality is that most will find it difficult to scale their AI pilot projects into enterprise level production, dramatically reducing the organisation’s ability to recognise the potential business value of artificial intelligence.

Chirag Dekate, Senior Director Analyst for Gartner says: “Although the potential for success is enormous, delivering business impact from AI initiatives takes much longer than anticipated... IT leaders responsible for AI are discovering “AI pilot paradox,”’ where launching pilots is deceptively easy but deploying them into production is notoriously challenging.”

The solution? AI must be carefully managed via ‘nurturing infrastructure strategies’ that sensitively facilitate AI pilots as they evolve, giving them space to grow into scalable production and so too, business value realisation. Coming back to PWC’s idea that putting the groundwork in with the basics is absolutely vital to future success.

So, what does a ‘nurturing infrastructure strategy look like? How can enterprise create the right environment for evolution?

Gartner says we must consider the following five predictions in the break-neck evolution of AI to successfully master production:

1. Infrastructure decisions will be driven by AI

Infrastructure decisions will be primarily driven by AI from no until 2023. By putting infrastructure resources front and centre of any AI pilot will allow the project to evolve alongside the technologies.

“AI models will need to be periodically refined by the enterprise IT team to ensure high success rates. This might include standardizing data pipelines or integrating machine learning (ML) models with streaming data sources to deliver real-time predictions.” Gartner
 

2. Increasing complexity of AI techniques managed through collaboration

Leveraging AI techniques like Machine Learning or Deep Neural Networks (DNN) in IoT environments faces many challenges, not least the complexity of data and analytics today.

Therefore, to successfully deploy production AI requires close communications and collaboration between ‘business’ and IT departments. This means proactively working to understand what drives the other, planning and providing “ready solutions when new business needs emerge” — a concept Gartner has coined ‘infrastructure-led disruption’.

3. Recognising that sometimes Machine Learning (ML) is smarter

From here until 2022 it is predicted that some 75% of organisations will use DNN’s where classic machine learning would more than suffice. Gartner suggest this is overkill.

Research shows that early adopters of ML already know this and successfully implement to deliver tangible business value. Those early to the party leveraged statistical ML in the first instance, carefully adding more advanced technologies (such as deep learning) and techniques only as the requirements arose, allowing them to realise the impact of AI first hand. The journey is as important as the solution here. 

Gartner says: “Sift through the AI hype and learn the spectrum of options to appropriately address business problems. Opt for simplicity over popular, but complicated, options.”

4. Cloud service providers should be part of your strategy

By 2023 AI will be one of the top cloud services, with use of and availability having increased fivefold since 2019. Not least because of the ability for cloud technologies such as containers,  cognitive APIs, and serverless computing to simplify the complicated process of deploying AI but also because they will enable machine learning models to serve as independent functions, reducing cost and overheads.
 

5. Adopt AI augmented automation

Embracing augmented automation in your AI will comfortably upskill your IT teams in the field while positioning them to better communicate with peripheral business units.

Organisations are dealing with growing levels of data and with that comes an increase in issues around problem prioritisation and false alarms. Problems in this area are compounded by the fact IT and business unit aren’t adept at communicating with each other, especially when it comes to the topic of AI.

Gartner says that by 2023, 40% of large enterprise Infrastructure and Operations teams will use AI-augmented automation that will result in greater agility, scalability and IT productivity.
 
 
 
 
 
 
 
 
 
 

Return to blog

We're Hiring

Interested and want to know more?Send us an email