Artificial intelligence is acquiring greater influence over how organisations distribute work, evaluate employees and make decisions. As intelligent systems become more deeply involved in everyday operations, business leaders face important questions about managerial authority, employee development and accountability.
By the time a manager opens a performance dashboard, an algorithm may have already determined which activities deserve attention and how an employee’s contribution should be interpreted. The system might have identified a decline in productivity, recommended changes to the employee’s schedule or assessed their performance against colleagues handling similar responsibilities. These recommendations can be useful, although their accuracy depends on the information collected and the assumptions used to interpret it.
Artificial intelligence is taking a growing role in these activities, gradually changing how decisions are made and where influence resides within organisations. Recruitment platforms can assess applications and recommend candidates for consideration, while intelligent workforce management systems predict staffing requirements and coordinate assignments. As these technologies become more capable, they are beginning to participate in activities traditionally associated with managerial judgement.
According to Stanford University’s 2026 AI Index Report, 88% of surveyed organisations reported using artificial intelligence in at least one business function in 2025. Meanwhile, McKinsey’s 2026 State of AI survey found that 44% of respondents reported scaling AI across their enterprises, compared with 38% the previous year.
As adoption expands, a different organisational model is beginning to take shape. In this emerging algorithmic organisation, intelligent systems participate in coordinating work, sharing expertise, evaluating performance and influencing business decisions. Their growing involvement has implications for how organisations establish authority, develop employees and hold decision-makers accountable for the outcomes of their actions.
How AI is changing the structure of organisations
Organisations have relied on technology to coordinate work for decades. Enterprise resource planning systems connect business activities, customer relationship management platforms track commercial relationships, and workforce management software helps businesses plan schedules and monitor performance.
These technologies established much of the digital infrastructure through which organisations collect information and manage operations. Artificial intelligence extends those capabilities by allowing systems to interpret complex information, identify patterns and recommend actions based on changing circumstances.
The development of agentic AI introduces another dimension. These systems can pursue defined objectives through connected applications, drawing on available information and completing sequences of tasks within established permissions. As they become more capable, businesses can delegate activities that previously required employees to coordinate several processes manually.
Recruitment offers a useful illustration. An applicant tracking system traditionally helps recruiters organise applications and manage candidates through the hiring process. AI-enabled tools can assess how applicants’ documented capabilities correspond with a role, summarise relevant experience and recommend candidates for further consideration. More advanced workflows can coordinate interview scheduling and prepare information for hiring managers.
These developments form part of a broader shift towards agentic AI coworkers, which can participate in business processes with increasing autonomy. Their responsibilities depend on the capabilities of the technology and the limits established by the organisation deploying it.
As intelligent systems become connected across business functions, their influence extends through the processes that determine how work moves between departments. A procurement system can assess supplier performance and initiate a purchase request, while connected financial applications examine expenditure against established budgets. HR systems can use workforce information to identify capability gaps and recommend development opportunities.
These arrangements introduce a different way of coordinating organisational activity. Information and recommendations can move through several intelligent systems before reaching the managers responsible for approving decisions. Executives must therefore understand how these systems interact, the authority assigned to them and the circumstances in which human intervention is necessary.
When AI becomes a colleague, expertise becomes more accessible
One of artificial intelligence’s most valuable contributions to workplace productivity is its ability to make useful knowledge available to employees when they need it.
In many organisations, expertise is concentrated among experienced professionals who understand internal processes, customer expectations and the judgement required to resolve difficult problems. Much of this knowledge develops through years of practice, making it difficult to transfer quickly through formal training.
AI-assisted systems can make elements of that accumulated experience more accessible, helping employees approach unfamiliar assignments with guidance informed by previous work.
A study by Erik Brynjolfsson, Danielle Li and Lindsey Raymond, published in The Quarterly Journal of Economics in 2025, provides evidence of this effect. The researchers examined the introduction of a generative AI assistant among 5,172 customer support agents and found that access to the technology increased productivity, measured by issues resolved per hour, by an average of 15%.
The improvements were particularly pronounced among less experienced and lower-skilled employees, who became faster and more effective at resolving customer problems. Highly experienced agents recorded much smaller productivity gains, while the researchers observed slight declines in quality among some of the strongest performers.
The study on generative AI at work suggests that intelligent systems can help distribute organisational knowledge that employees would otherwise spend considerable time acquiring. In customer-facing roles, this might include approaches to difficult conversations, methods for resolving unusual complaints and guidance on navigating complex internal procedures.
An employee encountering an unfamiliar problem can receive recommendations informed by previous interactions and benefit from experience accumulated across the organisation. Such assistance becomes particularly valuable during onboarding, when newly recruited employees are learning how to apply established procedures to unfamiliar situations.
IBM provides another illustration through its AskHR system, which helps employees access information and services relating to payroll, leave and other HR matters. According to IBM, the system automates more than 80 HR tasks and handles over 2.1 million employee conversations annually. Routine enquiries are managed through the technology, while human advisers handle matters requiring more extensive support.
For HR professionals, the arrangement can reduce the time spent responding to recurring administrative requests and create greater capacity for employee relations and complex workplace matters. It also changes the expectations placed on employees, who must learn when to rely on automated guidance and when circumstances require additional professional judgement.
As intelligent assistance becomes part of everyday work, organisations will need to ensure that employees continue developing the knowledge required to evaluate the recommendations they receive. The importance of that capability grows when algorithms begin influencing how their work is supervised and assessed.
When algorithms become involved in managing employees
The use of software to assign, monitor and evaluate work has developed over several years, particularly within logistics operations, digital platforms and businesses that coordinate large workforces. Its application now extends across conventional workplaces through systems that support scheduling, performance tracking and managerial decisions.
Commonly described as algorithmic management, this practice has become considerably more widespread than many organisations realise.
A 2025 OECD study on algorithmic management examined its adoption across six countries. The findings showed that 90% of surveyed managers in the United States reported that their firms had adopted at least one tool used to instruct, monitor or evaluate workers. The corresponding average across France, Germany, Italy and Spain was 79%, compared with 40% in Japan.
The technologies examined included conventional software alongside more advanced algorithmic tools, illustrating how automated processes have become embedded in managerial activities.
Their appeal is understandable. Managers oversee operations that require timely information and effective resource allocation. Intelligent systems can identify emerging problems, analyse performance patterns and help supervisors respond before operational difficulties become more serious.
A contact centre supervisor can receive early warnings about declining service quality or rising call volumes. In logistics, predictive systems can help managers adjust delivery assignments, while manufacturing businesses can identify equipment-related problems and direct technical support towards affected operations.
These capabilities become more consequential when information generated by intelligent systems influences employee evaluations. Organisations can collect detailed records of activity and use algorithms to interpret those records against established performance expectations.
Consider a recruitment consultant assessed primarily on the number of candidates submitted each week. An intelligent system can monitor submissions accurately and identify variations across the team. Yet a consultant handling difficult executive searches may produce fewer submissions while creating greater value through the suitability of candidates presented.
A similar challenge arises in customer service, where shorter handling times can appear desirable even when an employee needs additional time to resolve a complicated complaint. Within professional services, the value of an assignment may depend on careful research and sound judgement, qualities that are difficult to capture through activity-based measurements.
These situations expose a familiar weakness in performance management. Organisations often rely heavily on information that can be measured consistently, although the value of an employee’s contribution frequently depends on circumstances requiring qualitative assessment.
Algorithmic supervision can amplify this tendency because operational measurements become more visible and easier to incorporate into decisions about employees. The reliability of those decisions therefore depends on whether the measures provide a meaningful account of performance.
As explored in our article on how AI-assisted work should influence performance assessment, managers need to examine the quality of employees’ decisions and the contribution those decisions make to business outcomes. This becomes particularly important when intelligent systems perform a substantial share of the activities involved in completing an assignment.
The OECD’s research found that 60% of managers using algorithmic management tools believed the technology improved the quality of their decisions. However, nearly two-thirds expressed concerns about its potential effects on workers, with unclear accountability and difficulties understanding algorithmic recommendations among the issues identified.
For employers, the findings make the design of performance measures an important responsibility. Employees need to understand how information collected about their activities influences assessments, while managers require sufficient discretion to consider individual circumstances and contributions that extend beyond recorded activity.
When AI recommendations influence organisational decisions
The growing influence of artificial intelligence becomes more consequential when systems participate in decisions affecting employment opportunities, financial resources and individual welfare.
Machine learning systems can identify patterns across large volumes of information, helping businesses detect unusual financial activity, estimate future demand and assess operational risks. Generative and agentic systems extend these capabilities by analysing documents, comparing possible actions and coordinating activities across connected applications.
As confidence in these technologies grows, organisations may rely more heavily on their recommendations when making decisions about budgets, recruitment, employee development and customer relationships. The quality of those decisions depends on whether the systems have considered the information and circumstances necessary to reach an appropriate conclusion.
A recruitment system trained on historical employment records may identify patterns associated with previous successful hires. However, those records can contain the effects of earlier recruitment practices, unequal access to opportunities and inconsistent assessments of candidate suitability.
When an organisation accepts such recommendations without examining the underlying criteria, historical patterns can influence future hiring decisions. Candidates whose backgrounds differ from those represented in the training data may consequently receive less favourable assessments despite possessing relevant qualifications.
The consequences of automated hiring decisions are visible in a case involving iTutorGroup, an international online tutoring company.
In 2023, the United States Equal Employment Opportunity Commission announced a $365,000 settlement following allegations that the company’s recruitment software automatically rejected certain older applicants. According to the Commission, more than 200 qualified candidates were affected by screening rules that excluded female applicants aged 55 and above and male applicants aged 60 and above.
The case involved programmed recruitment rules, an earlier form of algorithmic decision-making, and demonstrates how software can translate organisational criteria into decisions with significant consequences for individuals. The EEOC’s account of the settlement also illustrates the legal responsibilities that remain with employers when discriminatory decisions are executed through technology.
As AI systems become more sophisticated, similar questions will arise across recruitment, promotion, compensation and workforce planning. Organisations need to understand which information influences their systems, how recommendations are tested and what procedures exist when an employee or applicant challenges an outcome.
The responsibility becomes more demanding when several intelligent systems participate in the same business process. An automated procurement workflow might assess supplier performance, initiate a purchase request and update financial records through connected applications. An error introduced during the initial assessment can influence subsequent activities before a manager becomes aware of the problem.
Business leaders therefore need sufficient visibility into how these systems interact and which actions they are authorised to perform. Clear decision rights and the ability to trace recommendations through connected workflows become essential safeguards as organisations delegate more activities to intelligent systems.
How accountability and employee trust influence AI adoption
Organisational accountability has traditionally depended on established reporting relationships. Managers oversee employees, functional leaders supervise business operations, and executive leadership is responsible for wider organisational performance.
As intelligent systems acquire greater influence over decisions, these structures need to account for activities carried out through software.
An employee receiving an unfavourable AI-assisted performance assessment should have a reasonable opportunity to understand the basis of that judgement. Where inaccurate information or unsuitable criteria have influenced the result, the organisation needs a process through which the decision can be reviewed and corrected.
Meaningful human oversight requires managers to have access to relevant information and sufficient authority to reach an independent conclusion. It also requires organisations to establish clear responsibilities for evaluating systems, investigating errors and determining when automated processes should be suspended.
These responsibilities extend across business functions. Technology teams need to understand the capabilities and limitations of the systems deployed, while operational leaders assess their suitability for particular activities. Legal and compliance functions must consider obligations associated with personal data, automated decisions and the rights of affected individuals.
Employee confidence will also depend on how these arrangements operate in practice.
AI-enabled systems can collect detailed information about task completion, communication patterns and employee performance, giving managers a more extensive view of how work is carried out. Such information may support better planning, although its use requires careful consideration when it influences decisions about individual employees.
Clear workplace policies should explain why information is collected, how it will be used and the circumstances in which algorithmic recommendations may affect employment-related decisions.
Employees also need opportunities to raise concerns when intelligent systems produce inaccurate assessments. Their experience of how managers respond to such concerns will influence the confidence they place in automated processes and the organisation’s ability to use them responsibly.
The changing role of managers and the future of professional development
As artificial intelligence takes on more operational responsibilities, managers will increasingly coordinate work performed through combinations of human expertise and intelligent systems.
This requires a broader understanding of team capacity. Alongside employees’ skills and availability, managers will need to understand which activities can be assigned to intelligent systems, how their outputs should be evaluated and where professional supervision is necessary.
A marketing manager developing a campaign might use AI to examine customer behaviour and review competitor activity before preparing an initial brief. The manager remains responsible for interpreting the research and determining whether the recommendations are appropriate for the organisation’s audience and commercial objectives.
The transition adds another dimension to the responsibilities already carried by middle managers. As discussed in our earlier examination of the pressures facing middle management in Nigeria, these professionals translate organisational strategy into everyday operations while managing performance expectations and employee development needs.
AI adoption will increasingly depend on their ability to explain changing processes, establish appropriate expectations and support employees as their responsibilities evolve.
The implications also extend to early-career development, particularly in professions where routine assignments provide opportunities to acquire practical knowledge.
Junior accountants develop judgement by reviewing financial records, young lawyers learn through document analysis, and recruitment consultants improve their assessment skills by examining applications and discussing candidates with experienced colleagues. As AI performs more of these activities, organisations will need to reconsider how employees acquire the professional knowledge required for complex responsibilities.
A 2026 World Economic Forum report developed with PwC found that more than one in three young workers globally were employed in occupations with medium to high exposure to AI-driven task change. The report highlighted the importance of redesigning early-career roles and strengthening the pathways through which employees develop professional capabilities.
The challenge deserves attention because professional judgement develops through experience. An experienced accountant can recognise unusual entries in a financial report because previous assignments have established an understanding of how the information should behave. A newly recruited employee receiving the same AI-generated analysis may lack the practical knowledge required to identify its weaknesses.
Organisations can address this through learning arrangements that encourage employees to examine AI-generated outputs, test alternative interpretations and discuss their reasoning with experienced colleagues.
As AI becomes embedded in daily work, talent development programmes will need to give greater attention to employees’ ability to interpret information and exercise judgement. Their effectiveness will depend on whether employees have opportunities to apply these capabilities within actual business processes, supported by appropriate guidance and feedback.
Why AI productivity gains must translate into business performance
Artificial intelligence has generated substantial expectations around productivity and cost efficiency. Yet enterprise research suggests that improvements experienced by individual employees do not automatically translate into stronger financial performance.
McKinsey’s 2026 State of AI survey found that 80% of respondents reported improvements in their individual productivity from AI use, while 37% attributed some positive impact on their organisation’s earnings before interest and taxes to the technology.
The difference raises an important consideration for executives assessing the value of AI investments.
A business may reduce the time required to prepare reports without improving the decisions those reports support. Customer service agents may handle more enquiries while recurring complaints continue to affect customer satisfaction. A recruitment team may process applications faster while struggling to improve the suitability of candidates eventually employed.
The financial value generated depends on how effectively organisations translate operational improvements into outcomes that matter to the business.
When AI reduces the time employees spend on routine activities, managers need to determine how that additional capacity will be used. It can create opportunities to improve service quality, strengthen customer relationships, undertake more detailed analysis or support business growth without equivalent increases in administrative expenditure.
These outcomes require changes to how work is organised and how employees contribute to business objectives.
Organisations also need to account for the full cost of introducing intelligent systems. Technology subscriptions, integration expenses, employee training, data security and ongoing quality assurance influence the financial returns available from AI investments.
For leadership teams, the evaluation of AI performance should connect technology deployment with relevant operational measures. Within customer service, these may include resolution quality and customer retention. HR functions can examine recruitment effectiveness and employee service outcomes, while finance teams assess forecast accuracy and processing efficiency.
Such measures help organisations determine where intelligent systems contribute value and where improvements to existing processes are needed.
What the algorithmic organisation means for Nigerian businesses
For Nigerian businesses, the development of algorithmic organisations presents opportunities closely connected to the country’s growing use of digital services and technology-enabled business operations.
Financial institutions, telecommunications companies, professional services firms, recruitment businesses and outsourcing providers already manage activities that depend on large volumes of information and recurring operational processes.
Within HR outsourcing, intelligent systems can support payroll enquiries, employee documentation and workforce reporting. Recruitment organisations can apply AI to candidate sourcing and the preliminary review of applications, with experienced professionals overseeing hiring decisions.
Contact centre operations offer another area of application. AI can help advisers retrieve relevant information, identify customer needs and improve the consistency of responses. For businesses serving international customers, these capabilities can contribute to more competitive service operations when supported by strong quality standards and effective employee development.
The effectiveness of these technologies, however, depends on the conditions under which they operate.
The quality of business records varies across Nigerian organisations, with some still depending on fragmented information and manual processes. Differences in language, communication styles and customer expectations can also influence the accuracy of AI-enabled interactions.
A speech-based customer service system, for instance, requires testing across the accents and language patterns encountered during actual customer conversations. Recruitment systems must be capable of interpreting local qualifications, employment histories and professional experience appropriately.
These considerations become important when businesses adopt systems developed using datasets and operating assumptions from other markets.
Nigeria’s regulatory environment also places responsibilities on organisations using automated decision-making systems.
Section 37 of the Nigeria Data Protection Act 2023 addresses decisions based solely on automated processing of personal data that produce legal or similarly significant effects on individuals. Subject to specified exceptions, the Act establishes protections concerning such decisions and requires safeguards that include opportunities for human intervention, expressing a point of view and contesting an outcome.
For employers, these provisions make it important to examine how automated systems influence decisions affecting applicants and employees, particularly where the consequences are significant.
Effective AI adoption will therefore require close collaboration between executive leadership, HR, technology and legal teams. Organisations need to understand the decisions their systems influence, the information they rely upon and the circumstances in which a human reviewer should intervene.
As deployment expands, internal reviews can examine whether these systems continue to perform as expected and whether employees have the knowledge required to supervise their outputs.
The growing availability of enterprise technology solutions creates opportunities for Nigerian organisations to modernise their operations. Sustained value will depend on the quality of the processes and management arrangements supporting those systems.
For businesses operating across international markets, the ability to demonstrate reliable governance and responsible use of AI may also become an important part of maintaining business relationships and meeting client expectations.
The leadership challenge of the algorithmic organisation
Artificial intelligence is changing how decisions move through organisations and the responsibilities of the people who manage them. As intelligent systems acquire greater influence over employee performance, operational activities and the allocation of resources, executives must consider how these capabilities fit within existing structures of authority.
The decisions made during this transition will influence how employees acquire expertise, how managers assess contributions and how organisations respond when automated recommendations produce unexpected outcomes.
For business leaders, this means examining the assumptions built into established management processes and determining how they should evolve as AI becomes more capable. Performance measures must remain relevant to the work being assessed, employees need opportunities to develop independent judgement, and consequential decisions require clear lines of accountability.
As intelligent systems become established participants in organisational life, the quality of leadership will depend increasingly on the ability to direct their capabilities towards meaningful business outcomes while maintaining appropriate oversight.
The organisations that succeed will be those whose leaders understand how work is changing and are prepared to adapt their management practices accordingly, ensuring that the growing influence of artificial intelligence is matched by the competence and responsibility of the people entrusted with its use.
