
Business is connected a never-ending quest to boost efficiency, trim costs, and summation productivity. Some of nan earliest known businesses -- ancient Mesopotamian traders -- inspired nan invention of writing. (Record keeping -- now that's a competitory advantage!)
Similar needs person existed successful each economical period. The large quality now is that AI exertion tin boost these efficiencies successful caller and exponentially profitable ways. Agentic AI is astatine nan halfway of this ratio boost.
According to Dan Priest, main AI serviceman astatine PwC US, "agentic AI refers to AI systems that tin autonomously perceive, decide, and enactment wrong a defined scope to execute goals, tin of collaborating pinch humans, systems, aliases different agents." (PwC, a.k.a. PricewaterhouseCoopers, is 1 of nan "Big Four" -- nan world's 4 largest master services firms.)
Also: 5 ways to beryllium a awesome AI supplier manager, according to business leaders
Agentic AI systems are different from nan erstwhile procreation of algorithmic business guidance systems we've been utilizing for nan past fewer decades. Agentic AI tin understand context, respond to changing situations without moving from a script, and activity toward defined goals autonomously.
Compared to accepted automation (and immoderate quality managers), agentic AI systems tin beryllium flexible, grip ambiguity, and make informed decisions astatine nan velocity of business operations. Agentic AI, Priest says, "helps organizations run pinch greater speed, intelligence, and scalability, fundamentally transforming really activity gets done and decisions are made."
Common barriers to AI integration
However, you can't elemental activity a magic wand and get enterprise-wide agentic AI that useful perfectly. There are galore challenges, including nan existing method indebtedness profoundly entrenched pinch bequest devices and processes, aversion to change, regulatory challenges, and deficiency of knowing and method AI skills wrong nan organization.
"Common barriers to achieving integrated supplier systems see fragmented information environments, deficiency of interoperability betwixt tools, and siloed organizational structures," says PwC's AI expert.
Ironically, nan implementation process itself tin inhibit successful AI adoption. Many companies commencement by pursuing an IT champion practice: implementing a caller strategy successful mini increments. Unfortunately, nan astir adjuvant AI systems thrive connected cross-organizational information, truthful nan stepwise attack often results successful fragmentation, inefficiencies, and pushback among stakeholders.
"Overcoming these challenges requires not conscionable exertion upgrades, but besides taste and operational shifts to let for cross-functional alignment and scalable integration," Priest explains. "Additionally, concerns astir security, compliance, and governance tin slow adoption, particularly successful regulated industries."
Also: The AI complexity paradox: More productivity, much responsibilities
To successfully deploy agentic AI enterprise-wide and acquisition its benefits, managers request to reevaluate business processes, create cross-functional coordination strategies, get afloat executive-level buy-in, and foster taste alteration passim nan organization.
The captious domiciled of proof-of-concept successful agentic AI
It's earthy for managers to initially beryllium reluctant astir giving up quality processes to a machine. However, nan cardinal to successful deployment is impervious of conception (POC). PwC's AI guru says, "POCs matter much than ever, particularly successful environments wherever skepticism still runs deep."
By initiating early-stage deployments that showcase nan benefits and soft modulation to AI-based operations, nan exertion itself tin show its effectiveness and benefits.
Also: 4 questions to inquire yourself earlier betting connected AI successful your business - and why
"The way from proof-of-concept to enterprise-scale AI starts good earlier nan POC itself," suggests Priest. "It originates pinch a smart strategy. Success hinges connected picking nan correct opportunities: high-potential, high-certainty usage cases wherever AI is well-positioned to present existent value. That early judgement call, wherever leaders are placing their bets, is what separates organizations that standard AI from those that stall out. When you take wisely, you group nan shape for a POC that isn't conscionable a trial of feasibility, but a objection of tangible business impact."
Naturally, location will beryllium failures astatine this stage. But nan cardinal is not misdiagnosing failures arsenic AI failures erstwhile nan guidelines origin tin beryllium traced to errors successful readying aliases strategy. Since POCs request to make existent worth early, beryllium judge to find ways to measurement that worth truthful that you tin move what mightiness beryllium claims of occurrence into tangible, measurably provable successes.
Achieving buy-in from nan group successful your statement
Achieving buy-in tin beryllium a challenge. One side-effect of improved ratio and agentic AI deployment is often a simplification successful occupation information for nan very stakeholders who mightiness champion specified a deployment. Although nan company's bottommost statement mightiness benefit, individual labor often fearfulness nan alteration associated pinch enterprise-wide AI adoption.
To antagonistic this concern, Priest advises business leaders to look for indications that squad members are consenting aliases enthusiastic astir being assisted by AI. He says, "Successful take hinges connected quality openness to utilizing it."
Also: AI agents will frighten humans to execute their goals, Anthropic study finds
Building spot successful AI agents hinges connected humans believing there's a meaningful worth proposition astatine nan extremity of nan AI journey. People request to spot clear benefits, whether it's efficiency, insight, aliases caller capabilities. Trust isn't conscionable astir performance, Priest says, "It's astir relevance. If users don't judge AI is moving successful their liking aliases delivering tangible value, skepticism will grow, sloppy of really precocious nan exertion is."
PwC's AI guru tells ZDNET, "We judge AI agents should beryllium utilized to empower people, not switch them. The ingredients required of a awesome squad are ones that AI agents are not capable to replicate, which see heavy specialization and expertise, diverseness of thought and opinion, and nan expertise to beryllium forward-thinking and creative."
He recommends that leaders hole their group for an AI-enabled future, which involves learning to activity alongside agents, to unlocking worth from data, to building high-performing teams wherever humans and agents collaborate to thrust innovation.
Also: How I personalized my ChatGPT conversations - why it's a crippled changer
AI agents tin augment nan workforce by taking connected routine, repetitive tasks, allowing labor to attraction connected much strategic, creative, and value-generating work. They tin service arsenic intelligent assistants by helping pinch tasks for illustration research, summarization, workflow automation, and determination making.
"That benignant of augmentation enhances productivity," Priese says, "While preserving nan quality judgement and discourse that machines can't replicate."
Practical examples of agentic AI successful action
PwC helps clients merge AI agents into their workforce strategies. When asked to place applicable occurrence stories, nan institution shared 3 examples successful technology, hospitality, and healthcare.
Technology: A awesome exertion institution reimagined customer engagement by deploying an AI agent-powered, omnichannel interaction center. With predictive intent modeling, adaptive dialogue, and real-time analytics, PwC says nan strategy reduced telephone clip by astir 25%, trim telephone transfers by up to 60%, and boosted customer restitution by astir 10%.
Hospitality: A ample hospitality institution streamlined guidance of its marque standards crossed its world portfolio by deploying agile workflows wrong a modern, AI-powered platform. Intelligent agents now automate updates, approvals, and compliance tracking, which has reduced reappraisal times by up to 94%.
Also: I'm an AI devices expert, and these are nan only 2 I salary for (plus 3 I'm considering)
Healthcare: A world healthcare institution transformed crab attraction by deploying agentic AI workflows crossed oncology practices. Intelligent agents streamlined objective and operational processes. They automated nan extraction, standardization, and querying of unstructured documents. This made it astir 50% easier for doctors and researchers to find useful objective accusation for precision treatments and studies. It besides drove a astir 30% simplification successful unit administrative load done AI-powered archive hunt and synthesis.
Building infrastructure and establishing governance
Infrastructure and governance spell manus successful hand. Agents, by their very nature, must recreation crossed organizational units and pass among disciplines and systems. As soon arsenic interoperability is introduced astatine that level, method compatibility becomes a awesome situation and requirement.
Standards, modular systems, and unfastened root implementations tin trim semipermanent risks and summation compatibility and maintainability. PwC recommends enterprises put successful scalable, unafraid platforms that support orchestration, observability, and integration crossed systems. This includes robust information pipelines, APIs, and governance frameworks to thief agents run reliably and responsibly astatine scale.
Also: What are AI agents? How to entree a squad of personalized assistants
"Effective governance frameworks for AI agents harvester clear accountability, robust oversight, and alignment pinch regulatory standards," says Priest. "Principles for illustration transparency, explainability, information privacy, and bias mitigation should beryllium embedded into some nan method architecture and organizational policies."
This is an ongoing process. Incorporate reviews, exemplary validation, and see human-in-the-loop mechanisms to thief support power arsenic agents scale.
The semipermanent outlook
PwC predicts that, complete nan adjacent 2 years, agentic AI will toggle shape really teams operate. Intelligence will go an intrinsic portion of business, starring to amended decisions, much informed leaders, and highly specialized experts.
"I'm excited astir this play because it marks nan opening of a high-performance era, wherever agents elevate teams to go nan smartest successful nan history of humanity," Priest says.
Also: Managing AI agents arsenic labor is nan situation of 2025, says Goldman Sachs CIO
Looking up 5 years, agentic AI will apt germinate into a foundational furniture of endeavor infrastructure. These agents will go progressively autonomous, tin of continuous learning, adapting to business goals successful existent time, and collaborating seamlessly pinch humans and different agents.
Priest tells ZDNET, "With these changes, it's important to retrieve nan large picture. The displacement we're experiencing isn't temporary, it's foundational and won't spell away."
What astir your organization?
Are you exploring agentic AI? Have you already begun integrating AI agents into your workflows? What challenges person you faced aliases do you expect erstwhile it comes to adoption, governance, aliases worker buy-in? Are location circumstantial usage cases wherever you deliberation AI agents could person a existent effect successful your business? Let america cognize successful nan comments below.
Want much stories astir AI? Sign up for Innovation, our play newsletter.
You tin travel my day-to-day task updates connected societal media. Be judge to subscribe to my play update newsletter, and travel maine connected Twitter/X astatine @DavidGewirtz, connected Facebook astatine Facebook.com/DavidGewirtz, connected Instagram astatine Instagram.com/DavidGewirtz, connected Bluesky astatine @DavidGewirtz.com, and connected YouTube astatine YouTube.com/DavidGewirtzTV.