Kathy Gibson reports from Pinnacle TechScape, Durban – IT distributor Pinnacle is making it easier for its partners and their customers to deploy artificial intelligence (AI) solutions.
AI might be the latest industry hype, but it is rapidly becoming a mainstream technology that adds value to organisations across the board.
Pinnacle’s AI Services organisation aims to help its reseller partners guide their end user customers in leveraging AI technology to improve their business outcomes.
Tim Humphreys-Davies, CEO of Pinnacle, tells delegates to Pinnacle TechScape that the AI market opportunity is significant.
The overall IT market is slowly returning to growth and exhibiting remarkable resilience. In EMEA, a growth rate of above 8% has seen IT spend reach $941-billion.
IT services still account for a significant share of spend at about 35%, with on-premise and cloud software enjoying above 20% growth and more than 35% market share.
When it comes to software vendors, Microsoft consumes the largest share of spend at about 14% to 18% – and growing at a healthy 15%. Coming off a lower base, AWS is growing at about 20%, and Google Cloud is growing at more than 30%.
“IDC tells us that the cloud laggards are going to be the AI laggards,” says Humphreys-Davies.
AI is very much the driving force behind growing IT spend, Humphreys-Davies points out.
Global studies indicate that large enterprises, on average, plan to spend $28-million on generative AI (GenAI) projects for a total market spend of $521-million – and an $11-trillion impact on the global economy.
“The trick for us and our partners is to have the right go-to-market,” Humphreys-Davies says.
At the moment, an $8-billiion EMEA spend is split between GenAI infrastructure, services, and platforms and apps – and these are all forecast to grow at close to, or above, 100%.
C-suite leaders have identified GenAI as one of the top strategic priorities closely followed by automation, cybersecurity, and direct-to-consumer platforms.
Jacques Visagie, GM: AI services at Pinnacle, explains that organisations recognise the importance of investing in GenAI, but not necessarily how to get started.
The usual starting point is with pre-trained, publicly-available models. These could be employed as is or with a company’s own data, or the organisation can go a step further and create its own models.
There is no single answer, Visagie stresses, with each use case standing on its own.
These use cases would typically include data analysis (according to 54% of business leaders), general research (50%), content creation (48%), teaching and or training (46%), software programming (43%), creating or improving personalised chat experiences (43%) and as a personal assistant (39%).
The main objective, Visagie adds, is for GenAI to enhance the work experience.
But, when it comes to deciding how to do this, end users and reseller partners are confronted with a multitude of AI platforms and models, machine learning frameworks, operating environments, software models, OEM vendors and silicon vendors.
Humphreys-Davies points out some of the challenges that partners have to overcome in order to bring AI projects to light.
The first – and arguably most important – of these is getting buy-in from the customer CFO. “You have to be able to measure the return on any AI investment, and it’s the first question customers are going to ask.”
Governance is also critical for any AI project. “Who will use the system and have access to the data? What can they do with it? What are the ethical boundaries? You have to get this past many milestones in the organisation.”
The big elephant in the room is an AI skills and talent strategy with the ability to hire – and retain – skills a critical consideration.
Long-term costs need to be considered – whether the AI investment is opex or capex, and what kind of long-term budgets will be required.
“In this environment, the way forward is to do the use case first,” Humphreys-Davies advises. “Look at what is a priority in the organisation and how GenAI can help to save costs. Follow this up with build and buy guidance, do governance by design, and make sure you have pricing predictability. It is vital that you have the AI-ready skills and that you retain them.
“So it’s less about build and buy, more about train and tune – with multiple options in between,” he says.
Visagie agrees that every AI journey is different.
Factors that need to be considered include what the data looks like, what skills are available, and which model would best suit the end objective. “Is there a model out there that pertains to my industry; should I work with a partner to transform and tune an available model; or do I need to build my own?
“But, at the end of the day, it doesn’t matter if you want to buy or build: It’s about the use case,” Visagie says.
Pinnacle AI Services helps partners to guide their customers by taking a service-led approach, consulting with customers to understand their use case.
“We will workshop with customers to understand their business and processes – and how AI can enhance workforce productivity.
“We will do an assessment across all verticals looking at what infrastructure, framework, and platform is in place.
“We will the provide guidance and capabilities to help the customer and partner to build an AI solution.”
Pinnacle will assist partners to do the proposal, to build the solution, and to run the project long-term.
“In fact, we have already invested in building this business and are differentiating ourselves with skills and capability,” Visagie says.
The distributor is currently building its own internal AI use case which Visagie believes will revolutionise the way it takes its products and services to market.
It has brought a data scientist and DevOps engineer on board and invested in consulting and enablement skills, leveraging resources in all of Pinnacle’s business units.
A warranty and support centre has been set up and Pinnacle will make a range of professional services available for its partners to white label.
Humphreys-Davies stresses that all of these services will be available to end user customers only via channel partners.
“We are a channel-centric organisation, so won’t sell services direct to end users,” he says. “As the AI market grows and partners scale, Pinnacle will adds resources and also transfer significant skills to its partner base.”