AI's business future: What's to come in the next 5 years
Five years can seem like an eternity in a world where AI's technological advancements occur almost daily, but best to be prepared now and avoid "future shocks."
Despite all the excitement about AI in the enterprise since the advent of ChatGPT for general use, businesses overall are still in the early stages of adoption. While it's a given that AI applications in IT and across entire business operations will continue to increase steadily in 2026, most businesses are still trying to ascertain the value propositions for various AI use cases.
As a result, most companies are exploring just a few use cases and waiting for other organizations to take the plunge on the majority of AI initiatives. Expect those initiatives to center on the adoption of AI-powered tools and processes across large swaths of business activities, such as low-hanging fruit like call center staffing, the upper reaches of business planning and supply chains that deliver goods and services.
As with any predictions about AI's future, some speculative and not-so-speculative assumptions are necessary, including the following:
- AI service providers and businesses will adopt more sustainable approaches to AI development, requiring numerous hardware, systems and software improvements. Chip and system designers are already creating GPUs and other AI-specific processors with dramatically lower power consumption. Using co-packaged optics and other photonics advancements, data center architects and engineers are shifting AI data centers to optical interconnects and networks, drawing less power and generating less heat while improving performance. Also, AI developers are continuing to find multiple ways of making training and inference less compute-intensive.
- A steady stream of small-scale AI-created problems is inevitable as use cases increase, and a major disaster is surely possible before the end of 2030. An AI-driven catastrophe could spur highly restrictive U.S. regulations that forestall planned AI investments.
- Large language models will continue to suffer from hallucination issues due to training errors or insufficiency. Safeguards are evolving, but it'll take more than five years to completely resolve.
- AI will continue to be plagued by security issues ranging from data leakage to insider weaponization. Bad actors will use AI more broadly to automate previously manual portions of their cybercrime workflows and disrupt AI deployments. An attacker, for example, might use voice bots to swamp customer service call centers as a denial-of-service or social engineering attack.
AI's greatest impact on businesses over the next five years will reach into several key areas: where AI is currently deployed; where businesses are eagerly waiting for AI to be sufficiently powerful, flexible and reliable; and where AI applications are only speculative and not widely discussed.
Contact centers
Chatbots in contact centers have been a source of customer frustration for years. Advanced call handling has been steadily improving at channeling customers into problem-specific queues, doling out FAQ answers and gathering more and better information from callers to optimize the time human agents spend after a handoff. Over the last year or so, the quality of AI-driven chatbots has improved to the point where call center operators can elevate AI to full call agents. Some contact centers are already downsizing human teams or hiring fewer contract agents. Expect most contact centers to have cheap AI chatbots as tier 1 support, more expensive agentic AI bots as tier 2 support, and small human staffs working as tier 3 escalation agents.
Client communications
One insurance carrier found that generative AI did a better job than most humans at communicating with insurance customers during the claims process. AI wrote clearer communications and was more empathetic with customers than humans. Expect AI agents to have more autonomy when communicating with customers.
Accounting and auditing
AI is automating a majority of financial auditing applications, including client expense reimbursements, telecommunications billing, accounts receivable and other complex financial transactions. AI agents don't suffer from fatigue, boredom or distraction, nor do they forget any policies guiding an audit. They can also spot patterns across vast volumes of transactions that a human might miss. AI already serves as auditor assistants, enabling human auditors to focus on questionable transactions. More companies will be embracing AI for internal and external account audits, with increasing levels of autonomy and with fewer humans needed -- or wanted.
Warehouse and logistics center staffing
Logistics centers and warehouses have experienced significant automation over the past several years using optimization algorithms and limited-function robots. This trend will accelerate and spread beyond large enterprises like Amazon and Target to midsize businesses and even small operators. AI will be able to optimize more kinds of operations and at price points that make it more affordable for a wider range of companies. Agentic AI packaged as general purpose warehouse robots will take over all item movements, especially container-packing, which will replace a good portion of human labor in midsize to large logistics facilities.
Delivery services
The accelerating adoption of self-driving cars in larger cities will be followed closely by self-driving delivery trucks. At first, these trucks will make business-to-business deliveries, such as warehouses to stores, since they'll require either humans or warehouse bots to receive and unload packages. Specialized delivery bots -- not necessarily general-purpose humanoid robots -- will eventually enable truck-to-door deliveries to consumers. Small businesses, however, will continue with human-only deliveries until services like DoorDash or Instacart take the lead on AI-driven robotic delivery.
Medium- and long-haul trucking
The benefits of automated medium- and long-range truck driving are compelling. AI won't get sleepy behind the wheel, take rest breaks, make side trips, violate traffic laws or stop for anything but gas and traffic. Expect fully and mostly autonomous trucks on some stretches of interstate by 2027 and possibly on some state highways in places like California and Texas. By the end of 2030, large shippers will do most of their shipping in corridors where these trucks are allowed.
Automated cargo ships
Using the agentic and physical AI technologies supporting autonomous vehicles and logistics robotics, automated cargo ships will employ skeleton crews of humans working mainly as troubleshooters for tasks like steering, loading and offloading the cargos. Some shippers might prefer to go without humans, opting instead to airdrop repair crews as needed, to avoid piratical attempts at kidnapping or attacking humans on board.
C-level management
The long-term trend has been to create more C-level positions that require greater human oversight and specialization. This trend could reverse as AI-driven automation makes sense in certain areas of C-level management, such as juggling the priorities and needs of various business lines and direct reports, especially in companies where mid-level management is significantly automated. Consequently, expect AI to drive the (re)consolidation of multiple C-level positions mainly in technologically aggressive companies.
Middle management
Businesses, developers and pundits often focus on AI replacing entry-level jobs and even higher tiers of customer-facing jobs. Less attention is paid to the possibility of AI replacing low-level and mid-level managers in office settings. AI might be able to organize and oversee workers more effectively than human managers when it comes to communication and report generation. AI doesn't have an ego, career ambitions or salary concerns. That said, upper-tier management might embrace the idea of handing over lower-level management responsibilities to AI. Expect technologically aggressive companies to deploy AI as line managers and further reduce the ranks of lower and middle management, even in areas where low-level jobs are mostly occupied by humans.
Project management
Most project manager responsibilities include tracking, organizing, and presenting information on a project's progress. Expect AI to be responsible for almost all the menial project management tasks in large companies, while human project managers focus almost exclusively on developing project plans. AI will assist with monitoring the entire project portfolio, predicting resource conflicts and troubleshooting.
John Burke is CTO and a research analyst at Nemertes Research. Burke joined Nemertes in 2005 with nearly two decades of technology experience. He has worked at all levels of IT, including as an end-user support specialist, programmer, system administrator, database specialist, network administrator, network architect and systems architect.
