Despite widespread AI adoption and rising investment, nearly half of mid-sized companies lack clear accountability for how employees and AI work together, while HR remains largely excluded from key workforce decisions, new research reveals.
Almost half (45%) of mid-sized businesses have no single person accountable for overseeing how employees and artificial intelligence (AI) work together, despite growing investment in the technology, according to new research from Lever talent. The findings reveal a significant disconnect between AI adoption and the value businesses believe they are achieving. Although 97% of surveyed companies have adopted AI and 92% are increasing their investment, just 28% describe the technology as extremely valuable to their business.
The Co-Intelligent Workforce Report: 2026 Baseline, based on a survey of 484 senior leaders at companies employing between 100 and 999 people across the US and Canada, suggests the challenge is no longer simply whether businesses are using AI, but whether anyone is taking responsibility for how it changes work. Only 21% of companies have someone whose primary responsibility is managing how people and AI work together. A further 34% have assigned the task to someone alongside other duties, while 12% report having no owner at all. Overall, 45% lack a single accountable leader.
“Every company is using AI,” said Drew Fortin, Founder and CEO of Lever Talent. “The question that matters now is what happens after the chatbot, and whether anyone in the building is responsible for the answer.”
The findings reinforce concerns raised in previous Fair Play Talks reporting on employees adopting AI faster than organisations can govern its use, highlighting the growing gap between technology adoption, workplace readiness and responsible oversight.
CLEAR ACCOUNTABILITY LINKED TO STRONGER AI RESULTS
Companies with dedicated leadership overseeing human-AI collaboration are significantly more likely to report stronger outcomes. More than half (55%) of businesses with a primary owner of their human-AI workforce describe AI as extremely valuable, compared with 22% of other companies. Among organisations where nobody owns the responsibility, that figure falls to just 10%.
Those with dedicated oversight are also more likely to maintain an up-to-date inventory of their AI systems and regularly review the technology’s impact on their workforce. However, the research suggests assigning responsibility is only part of the solution. Companies also need to establish which tasks AI should perform, where human judgement remains essential and who is accountable when automated systems make mistakes.
Where organisations have clearly defined both the division of work between people and AI and responsibility for correcting errors, 51% report extreme value from AI. That compares with just 9% where these arrangements are unclear. The need for stronger oversight is reinforced by previous research revealing that four in 10 US workers using AI have accepted answers they suspected were wrong. Together, the findings raise questions about whether employers have adequate safeguards to ensure AI-generated work is properly checked before influencing business decisions.
HR STILL EXCLUDED FROM CRITICAL AI DECISIONS
One of the report’s most striking findings concerns the limited involvement of HR in decisions about AI implementation and job redesign. Although HR departments are increasingly responsible for employee training, workforce planning and managing concerns about automation, just 18% of surveyed companies involve HR before selecting AI technologies.
When organisations appoint someone to oversee human-AI collaboration, the responsibility most commonly sits with IT, which holds the role in 48% of cases. HR is responsible in fewer than one in 10. Yet HR handles employee concerns about AI in 44% of companies and manages AI training in 41%, while helping determine how jobs should be redesigned in only 29%.
The consequences for business performance could be substantial. Some 58% of companies involving HR before AI tools are selected describe the technology as extremely valuable, compared with just 14% where HR becomes involved later or not at all.
The findings suggest that treating AI primarily as a technology investment, rather than a workforce transformation, could mean businesses overlook changes needed to make it effective. They also reinforce earlier Fair Play Talks reporting showing that three in four organisations lack a documented AI workforce strategy, despite HR departments facing increasing pressure to manage AI-related change.
TECH INVESTMENT ALONE IS NOT ENOUGH
The Lever Talent research also highlights significant differences between companies investing solely in AI technology and those taking a broader approach. Among businesses investing only in the technology, just 6% describe AI as extremely valuable. That figure rises to 50% among companies investing across technology, employee development programmes and people practices, including skills assessments and job redesign.
Yet 37% of surveyed businesses are still concentrating their investment on AI technology alone. The findings raise questions about whether employers are adequately preparing employees for changes to their roles, particularly as AI increasingly takes over administrative, analytical and routine tasks.
Previous global research found that most employers are preparing workers to use today’s AI tools rather than equipping them for tomorrow’s jobs. Separate IBM research revealed that six in 10 employees worry AI is weakening their skills, even as critical thinking and human judgement become increasingly important.
The findings suggest that investment in technology needs to be accompanied by meaningful opportunities for employees to develop new skills, adapt to changing responsibilities and maintain confidence in their ability to contribute.
WHO IS RESPONSIBLE WHEN AI GETS IT WRONG?
Clear accountability becomes even more important as businesses move beyond basic AI assistants towards autonomous systems capable of completing multiple tasks. According to Lever Talent, 61% of companies already using agentic AI cannot confirm that a human reviews every workflow before work progresses.
This creates potential risks around accuracy, decision-making and accountability, particularly where automated recommendations affect customers, employees or business operations. Previous Fair Play Talks reporting found that one in four executives say AI errors have already reached boards or external audiences, reinforcing concerns about the potential consequences of inadequate governance.
There are also implications for employees expected to check AI-generated work. The IBM research found that 43% of workers surveyed believe they receive the blame when AI makes mistakes, while 41% of Chief Human Resources Officers (CHROs) believe employees may not feel safe challenging AI outputs.
These findings suggest that organisations need more than technical safeguards. Employees must also have the confidence and authority to question automated decisions and raise concerns when something appears wrong.
THE COST OF INACTION: WHAT EMPLOYERS SHOULD DO
For employers, the research points to a need to move beyond simply introducing AI tools and take greater responsibility for how the technology is reshaping jobs, skills and everyday working practices. That means establishing clear leadership accountability, involving HR and employees from the outset and ensuring investment in technology is matched by investment in people.
Organisations also need to identify which tasks should remain under human control, where AI can genuinely improve productivity and who takes responsibility when automated systems produce inaccurate or potentially harmful outcomes. Training and employee development must be priorities. As AI changes the skills workers need, businesses risk leaving people unprepared if they focus on deploying technology without providing opportunities to adapt, develop and progress.
The consequences of inaction could extend well beyond disappointing returns on AI investment. Poorly managed implementation risks increasing employee anxiety, undermining trust in leadership, creating confusion over accountability and leaving workers facing additional demands rather than lighter workloads.
Indeed, past research has revealed that seven in 10 employees feel more productive using AI, yet their workloads are not necessarily falling. Workplace culture also matters. Earlier research found that half of Gen Z workers feel guilty using AI, while more than four in 10 admit hiding their use from employers.
RESPONSIBLE AI MATTERS
Without clear policies and supportive leadership, organisations risk encouraging employees to conceal AI use rather than openly discuss how the technology is helping them work. As behavioural scientist Dr Gleb Tsipursky previously argued in a Fair Play Talks guest column on AI transparency and psychological safety, responsible AI governance depends on employees feeling safe enough to disclose their use of technology, challenge questionable outputs and report mistakes without fear of blame.
Employers also need to consider whether AI is genuinely reducing work or simply transferring additional responsibilities to already stretched teams. Recent research found that more than eight in 10 senior leaders expect managers to remain as important or become more important in the AI era, even as their responsibilities for employee development, workforce transformation and AI oversight continue to expand.
Without clear safeguards, poorly managed AI implementation could also expose organisations to operational, reputational and, depending on how the technology is used, legal risks. Although Lever Talent’s findings reveal strong associations between leadership accountability, workforce investment and perceived AI value, they do not establish that these practices directly cause better outcomes.
Nevertheless, the wider evidence suggests businesses need to assess not only whether AI is delivering productivity improvements, but also how those improvements affect employees and whether responsibilities are being managed fairly and effectively.
LEADERSHIP WILL DETERMINE THE NEXT PHASE OF AI ADOPTION
Lever Talent argues that businesses need a more coordinated approach to managing their combined human and AI workforce. The company introduces the concept of a Workforce Orchestrator – a designated leader responsible for coordinating how technology, people and business processes work together.
Rather than leaving decisions about AI adoption scattered across departments, the approach is intended to create clearer oversight of workforce capabilities, technology investment, job redesign and business outcomes. For CEOs and HR leaders, the research suggests that the next stage of AI adoption will require closer collaboration between technology, HR and business leadership, alongside greater accountability for the impact of automation on employees.
The findings also arrive as more organisations confront difficult questions about the future of work, including which jobs will change, which skills will remain valuable and how employers can ensure that workers benefit from technological advances. The central challenge may no longer be how quickly businesses adopt AI, but how deliberately they redesign work around it. As AI becomes embedded in everyday working life, the real test of leadership may be less about how quickly companies embrace the technology and more about whether they can demonstrate that it delivers value without leaving their people behind.
Click here to read the full report from Lever Talent.
RELATED ARTICLES
Employees are embracing AI faster than many organisations can govern it, according to new research.
More than eight in 10 leaders say managers’ roles have grown over the past three years, while only one in 10 believes AI will reduce their importance, new research finds.
Why responsible AI governance must protect workers from blame while helping organisations build trust, accountability and responsible AI governance.
Half of Gen Z workers feel guilty using AI and more than four in 10 are using it without their employer’s knowledge, despite businesses increasingly demanding AI skills from new recruits.
Seven in 10 employees say AI is making them more productive, yet workloads remain virtually unchanged, reveals new research.
Six in 10 employees worry AI is weakening their skills, while 43% say they get the blame when AI goes wrong, new global research reveals.
HR teams are buckling under mounting AI pressures, with 95% of HR leaders reporting rising workloads while three in four organisations still have no AI workforce plan, according to new UK research.
