Workplace relationships hit hard by AI adoption - here's the HR playbook
Generative AI is changing how people talk, decide, and get work done. It's also straining relationships at work. New global findings from The Adaptavist Group show a clear trend: employees are swapping human interactions for AI prompts, and the ripple effects are already showing up in culture, communication, and skills.
For HR, this isn't just a tech rollout. It's a people strategy moment. The goal: keep the gains from AI without letting it erode trust, etiquette, and capability.
What the data says
- 32% of employees say they speak less to colleagues since adopting generative AI.
- 48% consult AI on legal or policy issues; 41% use it for HR-related matters.
- 26% would rather make small talk with an AI bot than a human; that climbs to 32% for business leaders.
- 26% admit they've become less polite since using generative AI.
- 29% worry their abilities are declining because they rely on AI.
Why employees are turning to AI
"As GenAI continues to embed itself in society, we're seeing a shift in how work gets done and how people connect and communicate," said Neal Riley, AI innovation lead at The Adaptavist Group.
CyberPsychology specialist Carolyn Freeman points to the appeal: AI is always available and non-judgemental. "Interactions with AI can feel more private, contained, and forgiving… giving employees space to vent, explore ideas, or rehearse difficult conversations without fear of damaging performance reviews or workplace reputations."
Her warning is clear: over-reliance can displace the real human contact people need to build trust and effective teams.
The opportunity if you set the right culture
Riley stresses balance: "The key to achieving a culture where AI use remains healthy and work-conducive is one which encourages responsible AI experimentation, and has a framework for measuring success."
Where organisations actively encourage responsible experimentation with AI, they report:
- Growing skills in AI-friendly organisations (89%)
- Improved satisfaction (54%)
- Better team collaboration (68%)
- Successful ROI on AI (73%)
In short: the right culture boosts performance, clarity, and measurable returns.
Risks HR can't ignore
- Communication drift: Less human contact weakens team cohesion and psychological safety.
- Etiquette erosion: Reports of lower politeness point to tone and empathy slipping in fast, AI-assisted workflows.
- Skill atrophy: Over-dependence can reduce critical thinking, writing, and judgement.
- Risky advice-seeking: Employees using AI for legal/policy or HR advice can create compliance, privacy, and equity issues.
The HR playbook: make AI use healthy and work-conducive
- Set guardrails: Define use cases that are allowed, restricted, or banned (e.g., no AI for legal decisions, health data, performance ratings, or sensitive employee issues).
- Create an AI etiquette code: Require disclosure when AI drafts content; set standards for tone, empathy, and context checks before sending.
- Train for judgement, not just prompts: Teach fact-checking, bias review, and red-teaming. Make "proof before publish" a habit.
- Offer safe channels: Provide approved tools with logging, data controls, and clear privacy notices.
- Protect human time: Block focus hours and schedule regular team connection rituals (standups, retros, office hours) to keep relationships strong.
- Reroute sensitive queries: Steer legal/policy questions to counsel or compliance; steer HR questions to HRBPs or service desks.
- Measure and tune: Track the metrics below; adjust policy, training, and tooling quarterly.
- Manager enablement: Coach leaders to model healthy AI use and to spot early signs of disengagement or skill slippage.
Metrics to track
- Connection: Manager 1:1 frequency, team meeting cadence, and cross-functional touchpoints per person.
- Quality: Peer-review pass rates on AI-assisted work; tone/politeness scores from internal surveys.
- Capability: Writing and problem-solving assessments over time; percentage of work sent without human review.
- Risk: Number of AI use exceptions (e.g., legal/HR queries to public models), privacy incidents, and flagged outputs.
- Value: Cycle-time reductions, ticket deflection, and project ROI tied to AI-enabled workflows.
Policy pointers and frameworks
- NIST AI Risk Management Framework - useful for structuring risk controls and measurement.
- ICO guidance on AI and data protection - practical guardrails for privacy, fairness, and transparency.
Quick start (next 30 days)
- Publish a one-page AI acceptable use policy and etiquette guide.
- Stand up an AI review board (HR, Legal, Security, Data) with a simple intake form.
- Roll out a 60-minute training on safe, effective prompting and fact-checking.
- Enable one approved AI tool with enterprise controls; disable unapproved browser plugins.
- Add two weekly human-connection rituals per team (e.g., standup + problem-solving huddle).
Bottom line
AI can speed work, but it won't build trust for you. HR's job is to set the conditions: guardrails, skills, human connection, and clear measures of success. Do that, and you'll get the upside without losing the culture that keeps teams strong.
Want structured, job-focused upskilling for your teams? Explore practical programs at Complete AI Training.
Your membership also unlocks:
