In 2025, experts say businesses will be vigilant while still increasingly employing AI in their services.
Generative artificial intelligence (GenAI) has proven its capability as a vital tool in nearly every industry, with Accenture recently finding that the technology could potentially add up to €148bn to Ireland’s annual gross domestic product by 2038.
S&P Global Market Intelligence data reported that GenAI start-ups secured more than $20bn in just the first three-quarters of 2024, highlighting the interest business have in the evolving technology. While according to the International Data Corporation, investments in AI are expected to near €700bn by 2028.
The technology has had a marked impact in every major industry, including in cloud services, automating repetitive tasks that detect and rectify errors and helping maintain security – a key concern for businesses, considering the mounting cybersecurity risks which can be augmented with the increased implementation of AI.
According to Forrester’s analysis, while both public and private cloud AI offerings will grow in 2025, challenges such as supply shortages, quality concerns and data security will impact market behaviour.
Here is what some experts believe will impact AI, cloud and data services this year.
Resolving errors with RAG
Responding to hallucination and accuracy errors in GenAI models, cloud players will shift their focus to retrieval-augmented generation (RAG), a technique that enhances the accuracy and reliability of GenAI large language models by filling the gaps in its functioning through external sources.
Forrester principal analysts Lee Sustar and Tracy Woo say that “in the year ahead, every major hyperscaler will launch RAG-related solutions to build accuracy into their GenAI services”.
Moreover, a Forrester reporter from last year identified that using a RAG approach can protect businesses and customers through its increased accuracy.
Enhanced focus on data protection
Data protection has taken mainstream importance with businesses far and wide suffering from increased cyberattacks and data breaches in recent years.
Cloud players, who deal in data, will need to be increasingly vigilant around who gains access to their data.
Jakob Østergaard, the chief technology officer at SaaS data protection business Keepit, explains that this year, businesses will see data protection as beyond an option and embrace it as a “non-negotiable priority”.
“Companies will demand solutions that not only secure their data but also guarantee accessibility under any circumstances.”
Moreover, Ricardo Batista, the managing director for UK and Ireland at Noesis, a tech consulting company, believes that data governance – or the act of ensuring data security and accessibility – will assume a pivotal role within organisations.
“Such measures are essential to ensure comprehensive control and security in this rapidly evolving landscape.”
This year, organisations will scrutinise their cloud service providers with the same “rigor” they apply to physical supply chains, explains Østergaard.
“Compliance, vendor relationships, and security protocols will come under the microscope…the days of blind trust are over.”
Forrester predicts that cloud specialist vendors such as Cisco and Palo Alto Networks will become considerable competition for hyperscalers such as AWS, Azure and Google.
“By the end of 2025, 60pc or fewer cloud customers will prefer the hyperscaler platform’s native Cloud Workload Security capabilities (CWS), while the remaining 40pc will use a platform specialist CWS vendor.”
Despite risks, more AI agents
Businesses have been increasingly employing AI chatbots and agents that are taking over manual and repetitive roles previously held by humans.
Max Ball, the lead author of a Forrester report on conversational AI, spoke with SiliconRepublic.com last year and explained that despite the challenges associated with customer-facing AI bots, which include data privacy risks, compliance with the constantly evolving regulations, hallucinations and a general lack of trust in AI, businesses are increasingly moving towards employing the technology.
The evolution of AI technology has brought advancements to AI agents, which according to Batista, “represents a paradigm shift that redefines the limits of traditional artificial intelligence”.
“With a high degree of autonomy, these agents are capable of carrying out predefined tasks independently, revolutionising multiple areas.
“More than having just a chatbot replying to some questions or doing some actions, AI agents will be talking between them, orchestrating actions and proving a task-oriented process where actions will be done by agents and supervised by humans.”
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.